Afghan Refugees

Finding Durable Solutions in Contested Transitions
The Case of the Afghan Refugee Population

Paper presented at the 49th Annual Convention of the International Studies Association
San Francisco, CA, 26-29 March 2008 (on Saturday, 29 March 2008, SA53)
Susanne Schmeidl (Griffith University)

and William Maley (Australian National University)1
Introduction
After 30 years of turbulence and vast repatriation efforts, Afghanistan still tops the
worlds list of sources of forced migration. At the height of the refugee crisis, Afghans
accounted for roughly 60% of the entire world refugee population. Mobility has become
not only a key livelihood and survival strategy for many Afghans, but an integral part of
their lives. Possibly two out of three Afghans have been displaced at least once; families
have scattered across countries and continents.
After an overview of the political/military context and resultant displacement, this
chapter traces the current situation of return in a country undergoing a political transition.
In addition to the two neighbouring countries of Iran and Pakistan, Western countries
also push for voluntary return in response to domestic concerns regarding asylum
seekers, and as a sign that all is not yet lost in Afghanistan. The use of voluntary in this
context, however, is problematic, especially when the bulk of the remaining refugees
have lived most of their lives in exile and many have never set foot in their home country.
Many returnees are caught in makeshift camps, squatter settlements, or ad hoc
accommodation and they face renewed internal displacements due to insurgency
violence, landlessness or natural disasters such as floods. Failures of physical,
economic, social and political reconstruction, unfortunately, do not provide fertile ground
for resolving the Afghan refugee situation any time soon. The current deteriorating
security situation in Afghanistan, coupled with insufficient humanitarian access to vast
parts of the country for the United Nations High Commissioner for Refugees (UNHCR)
and other international actors, raises concerns about protection. If the situation spins
further out of control, the country will be faced with yet another wave of forced migration.
We will examine the sustainability of voluntary return as a preferred durable
solution for Afghan refugee communities against the backdrop of national and regional
political and security considerations. Increasingly, Afghan refugees have become a
political football for the regional powers; the two major host countries (Pakistan and Iran)
have made the return of Afghan refugees part of their regional as well as international
policies. The transitional nature of the Afghan refugee situation, in the end, may show
that voluntary return is anything but a durable solution at this juncture and that
sustainability may not equate returning to places of origin, highlighting the need to
1
The authors would like to thank Alessandro Monsutti, Pierre Centlivres, Micheline CentlivresDemont, Salvatore Lombardo, Ewen McLeod and Ceri Oeppen for helpful comments on this
chapter. This paper will be revised for inclusion in Howard Adelman (ed., forthcoming). No Right
and No Return. Ashgate Publishers. Please direct all questions and queries to Susanne Schmeidl
at susanne(at)schmeidl.com
2008 ISA Presentation (Schmeidl, Maley)
Page 1 of 44
begin exploring alternatives for much of the remaining refugee population. Sustainability
of return needs careful consideration in light of the limited absorptive capacity of
Afghanistan.
Overview of the Context
Demographics and division
Presently, the Afghan population is estimated at something over twenty million, but
no comprehensive population census has ever been carried out (with the last one in
1979 still being the most accurate one). The population is segmented on ethnic,
sectarian, geographical, occupational, and life-experience lines. Ethnographers have
identified more than 50 ethnic groups (Orywal 1986, Schetter 2003, Jawad 1992) but
there are four dominant ones: ethnic Pashtuns and Tajiks are by common consent the
two largest groups with Hazaras and Uzbeks tied for third place at (Glatzer 1998). Within
these ethnic groups there are further divisions. Pashtuns are divided into two big tribal
confederations (Ghilzai and Durrani) with additional tribal sub-divisions (Glatzer 1998,
2001, 2002). The majority languages are Pashtu and Dari (Afghan Persian), but many
minority languages also prevail.
While almost all Afghans are Muslim, the population reflects the familiar SunniShiite division, with around 80% Sunni and the rest Shia. Within these groups there are
significant further branches, such as Twelvers (also the dominant sect in Iran) and
Ismailis among the Shia. In addition, ideological or attitudinal differences exist, notably
between intellectual and village Islam (Saikal and Maley 1991).
There was, and still is, also a very noteworthy difference in orientation between
rural dwellers and the residents of cities: the lives of the former tended to revolve around
cyclical forms of economic activity associated with agriculture or pastoralism, while the
latter were more likely to be involved in secondary or tertiary industries, or to be
employees of state bureaucracies. There are also Kuchi nomads who may total
anywhere from 600,000 to three million. The Afghan wars, but also years of drought,
diverted the rural and nomadic populations towards semi-urban and sedentary lifestyles
in refugee camps and later upon return home.
Historical Background
In April 1978, a communist coup led to a major slide to war, aggravated by the Soviet
Invastion of Afghanistan in December 1979. By the beginning of the 1990s there were
an estimated 6.2 million Afghans living as refugees outside the borders of their
homeland (Colville 1997) out of a pre-war population estimated at just over 13 million
(Eighmy 1990, 10). This refugee population was not created by steady accretion but
paralleled the phases of conflict (Goodson 1998). Though some Afghans became
refugees prior to the Soviet invasion in 1979, millions moved to Pakistan subsequently.
As large numbers returned following the collapse of the communist regime in April 1992,
still others fled Kabul as rival mujahideen factions bombarded the city. The emergence of
the Taliban from 1994 created new incentives for flight, while others returned hoping for
Page 2 of 44
an end to arbitrary and fragmented rule (see Figure 1). The overthrow of the Taliban in
late 2001 led to some additional forced migration, more than offset in 2002 when nearly
2 million refugees returned home (UNHCR 2007b).
Figure 1. Afghan Refugee Population, 1979-2006

Total
Pakistan
Iran
Elsewhere
7,000,000
6,000,000
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
20
05
20
03
20
01
19
99
19
97
19
95
19
93
19
91
19
89
19
87
19
85
19
83
19
81
19
79
Source: UNHCR (Statistics of Population of Concern, various years)

Displacement over time has varied in its precise form and character. Many
Afghans have been internally displaced, some on numerous occasions. Others became
refugees in neighbouring states, in some cases as refugees sur place (that is they
became refugees when they were outside Afghanistan), especially in Iran, and on other
occasions as either anticipatory refugees who fled in the early stages of upheaval with
some resources or acute refugees who left with little forewarning, mainly for Pakistan
and able to take only a few possessions (Kunz 1973). Smaller numbers moved on to
countries of resettlement, adding substantially to the Afghan migr communities in the
United States, Germany, Australia, India, and the Russian Federation. With the passage
of time, the consolidation of Afghan communities abroad, family reunification and
marriage have led to new patterns of chain migration. During the late 1990s as the
Taliban movement expanded in Afghanistan, some Afghans used smuggling networks to
move abroad, fearing that adjacent countries would be increasingly inhospitable (Maley
2001).
Page 3 of 44
Not all flows were necessarily driven purely by push factors. Some Afghan
communities had strong traditions of mobility to access labour markets in nearby states
(Monsutti 2005, Hanifi 2000). For others, the flight of the Prophet Muhammad from
Mecca to Medina in 622 CE, known as the Hijra, provided religious backing for flight in
face of threats; indeed, the word muhajirin, commonly used amongst Muslims as a label
for refugees, derives from the same Arabic root as the word hijra itself, and imports a
rich range of positive connotations that the English word refugee does not (Shahrani
1995, Centlivres and Centlivres-Demont 1988c). While reasons for flight were extremely
complex (Guillo et al. 1983), and survey evidence points to a range of specific factors
triggering or underpinning refugee movements (Connor 1987), the main cause of the
Afghan refugee outflow in the 1980s was fear of physical insecurity. The bulk of
movements to Pakistan (Sliwinski 1989a) came from Afghan provinces near the Durand
Line (the boundary separating Pakistan from Afghanistan drawn by Sir Mortimer Durand,
in 1893 that divided the Pashtuns between the two territories) as the USSR sought by
force to secure their clients from the threat of resistance coming from mujahideen
fighters operating from refugee camps in Pakistan. Figures on mortality in Afghanistan
during the period of Soviet occupation underscore the rationality of flight as a strategy,
as unnatural deaths in Afghanistan between 1978 and 1987 amounted to 876,825, or on
average over 240 every day throughout the decade (Khalidi 1991).
Regional political and security dimensions
In the complex regional geopolitical situation, refugee populations serve as tools
of neighbouring power interests. Pakistan, which after 1979 became the host for the
largest single population of Afghan refugees, historically had an extremely tense
relationship with Afghanistan (Grare 2003). Prior to the 1947 partition of India, the
Afghan government helped foster claims of a right to self-determination for Pashtuns in
India under the rubric of a demand for a distinct unit to be called Pashtunistan.
Afghanistan was the one state to vote against the admission of Pakistan to the United
Nations From the 1940s to the 1970s, the issue flared from time to time, leading to a
severing of diplomatic relations between Pakistan and Afghanistan between 1961 and
1963 (Maley 2002, 68-69; see also Qassem 2007).
The April 1978 coup detat, internal strife, and subsequent Soviet invasion to shore
up communist rule under a reliable puppet triggered a deep chill in East-West relations
(Westad 2005, 299-330) and provided Pakistan with an unprecedented opportunity to
reshape its strategic environment and become a frontline state, placed for the first time
in its history at the top of Americas strategic regional priorities (Grare 2003, 67).
Afghanistan became the site of the biggest proxy war between East and West in the last
decade of the Cold War, with Pakistan facilitating financial support to the Afghan
resistance. The historic Pashtunistan dispute strongly disposed Pakistan to support
more religiously radical Afghan resistance groups, seen as less likely to revive a
territorial-nationalist dispute in the future.
Pakistans involvement in supporting radical groups was somewhat blunted by the
events of 11 September 2001, but not eliminated. Sanctuaries in Pakistan continue to
Page 4 of 44
sustain an active neo-Taliban movement responsible for insurgency in southern

Afghanistan. Indeed, in August 2007, Pakistani President Musharraf, speaking in Kabul,
stated that The problem that you have in your region is because support is provided
from our side (Maley 2007, 5).
While this positioned Pakistan as a proactive regional actor in Afghanistan, other
regional states had interests as well. Iran assumed a certain protective responsibility
with respect to Afghanistans Shiite Muslim minority. In turn, fear that Afghanistan could
fall under Iranian influence led states such as Saudi Arabia to support Sunni Muslim
groups based in Pakistan. Individual Saudis, such as Osama Bin Laden, made their way
to Peshawar in Pakistan where they supported Arab volunteers associated with radical
Sunni resistance groups (Rubin, 1997). Thus, Afghanistan became a theatre in which
diverse rivalries - initially between the Soviet Union and its opponents, and then between
states reflecting different strands of Islamic thinking - were played out.
War also distorted traditional lifestyles. Afghans inhabit simultaneously a
multiplicity of social worlds, and circumstances may well influence which of these worlds
is most powerful in shaping their responses to particular situations. Further, the
importance of specific social cleavages is not fixed but variable, and this has particular
pertinence to the prospects of securing durable solutions to refugee problems. For
example, in regions in which ethnic differences were not exceedingly salient, recruitment
practices of mujahideen factions (and later the Taliban) along ethnic (and also tribal)
lines escalated latent group tensions into conflicts or created new ones altogether. While
Afghanistan has not experienced mass ethnic conflict, there have been some grim
specific episodes that have left their mark on the psyches of different refugee groups.
The Taliban, overwhelmingly from the Pashtun ethnic group, engaged in a number of
massacres of ethnic Hazaras, who were not only distinctive in their physical
appearances, but also largely Shia and thus exposed to charges of heresy from more
extreme Taliban circles. On the other hand, Pashtuns from the North, descendants of
settlers re-located there in the late 19th century (Tapper 1983), faced harassment and
forced expulsion during the short rule of the mujahideen government (1992-1996) when
the Uzbek warlord Abdul Rashid Dostum controlled much of the region. This continued
after 2001 when evidence surfaced that these Pashtuns fell victim to revenge attacks
from Taliban victims who saw them as having been complicit in earlier Taliban
domination out of ethnic solidarity (International Crisis Group 2003).
Socio-economic and resource contexts
Afghanistan has long been one of the poorest countries in the world, and across a whole
range of socio-economic indicators such as income per capita, life expectancy at birth,
and maternal mortality, its performance is little short of catastrophic (Maley 2006, 79).
Strong norms of reciprocity and social solidarity have historically helped to circumvent
major crises such as widespread famine (except in the early 1970s), but the livelihoods
of many Afghans are under constant pressure. Day-to-day survival depends on
engaging in a multiplicity of economic activities, and occasional borrowing. Large
numbers of Afghans have no defensible legal title to their property, and are unable to
Page 5 of 44
use it as collateral for borrowing; thus, much of the countrys wealth cannot be used to
facilitate productive investment. Furthermore, the turmoil of the last thirty years has led
to serious deterioration in infrastructure and capital, loss of skills, and major damage to
the states capacity to enforce contracts.
Although the Afghan economy has never collapsed, the disruptions since 1978
have handicapped significant elements of the economy with little in the way of a banking
system, increasing the significance of hawala informal money transfers (Monsutti 2005);
poor infrastructure (creating local autarkies with consequent loss of potential gains from
access to wider regional markets); and illegal opium crops. Afghanistan may be or
become a narco-state as the worlds largest opium producer (8,200 tons in 2007),
reflecting a range of opium economies with differing incentive structures at work
(UNODC 2007). This requires a range of nuanced responses that take into account local
specificities (Mansfield 2007).
Given the fragility of the Afghan economy as well as deteriorating security, the
Afghan government has recently discouraged further repatriation, fearing disastrous
consequences on top of Kabuls enormously swollen population of over 4 million (one of
the fastest-growing conurbations in the world). This has undermined the January 2006
Afghanistan Compacts provisions for refugee repatriation, and raises the question of just
how sustainable return is likely to be.
Stakeholder Analysis Roles, Interests and Approach
Stakeholders include the diverse Afghan refugees who are active agents in
shaping their futures, the United Nations High Commissioner for Refugees (UNHCR)
involved in refugee assistance and return, the two main countries hosting major Afghan
refugee populations - Iran and Pakistan, and international actors with regional interests.
Support for Afghan refugees from Western countries, especially the US, was
largely driven by a Cold War mentality of curbing Soviet expansion in line with the Carter
Doctrine in the early 1980s and a much more forceful Reagan Doctrine in the mid-1980s,
pursuant to which the US wanted to maintain a dominant position in the Persian Gulf
region (Schch 2008). Donor fatigue began to set in after the Soviet Union withdrew
from Afghanistan in 1989 and the Cold War ended. Following the emergence of the
Taliban, which the US initially tolerated, the US relinquished its weighty position in the
region for a period of time (Gutman 2008), something which it came to regret given its
weak position when the 9/11 attacks occurred. The USs great interest in Afghanistan
was subsequently revived, and prompted the invasion by Coalition Forces (under US
leadership) to overthrow the Taliban for having hosted Al Qaeda (Schmeidl 2007). Since
its war on terrorism started, the US has seen Afghanistan, and especially Pakistan, as
major allies.
After the Bonn Agreement in December 2001, the International Security Assistance
Force (ISAF) with 26 contributing nations in Afghanistan entered the arena but was
initially blocked from expanding beyond Kabul. The US introduced Provincial
Reconstruction Teams (PRTs) at the end of 2002 to add a reconstruction element to the
war on terrorism (Stapleton 2007a), and ISAF was put under NATO command and co2008 ISA Presentation (Schmeidl, Maley)
Page 6 of 44
ordination in August 2003. There are currently 25 PRTs in Afghanistan, 11 led by the US
(the largest donor country) and others led by various contributing nations (Maley 2006).
Refugee Organisations
Afghan refugees have had dynamic social and political lives, albeit constrained by the
unnatural milieu of the refugee camps in Pakistan, by broader resource scarcities, and
by particular problems for women refugees (Dupree 1992, Centlivres-Demont 1993). Far
from camps constituting an apolitical humanitarian space (Baitenmann 1990), they
became important venues for political interests where different Afghan actors struggled
to achieve superordinate positions, and for nurturing refugee warriors. The overall
political agenda was displacing the Soviet Union and its Afghan communist surrogates
from positions of power (Centlivres and Centlivres-Demont 1988a, Terry 2002, 55-82;
Lischer 2005, 44-72).
Refugee organisations mirrored Afghan traditional society where each village or
community (qaum) usually had either appointed (malek) or inherited (khan) traditional
leaderships that represent their quams to outsiders and the central government. Due to
disruptions during displacement, and to some leaders having joined the mujahideen,
new leaders, rupiah maleks, emerged who grew powerful mainly through controlling aid
and the distribution of food (Centlivres and Centlivres-Demont 1988a; Centlivres and
Centlivres-Demont 1988b; Centlivres 1993). Maleks were chosen by communities living
in the refugee camps in part for their organizational and linguistic skills (such as
speaking Urdu or English - Centlivres 1993, 28). Another traditional-oriented refugee
organisation revolved around conflict resolution through appointed committees of
influential individuals (Centlivres 1993, 28) that mimicked the traditional conflict
resolution mechanism of a jirga or shura in Afghanistan. If conflict involved outsiders,
new leaders with technical and linguistic skills were called to the job.
Refugees in Pakistan were required to register with one of the seven Sunni
mujahideen parties which Pakistani authorities sanctioned (Schch 2008). Thus, political
parties of the Afghan resistance (Centlivres 1993) served as intermediaries in dealing
with the refugees, mirroring the role of Afghan government officials such as the arbab,
especially in dealing with Pakistani officials or in helping refugees secure access to
official government relief. As a direct result, resistance parties were legitimised through
enforced support. Furthermore, several refugee camps were under the control of
commanders of the Afghan resistance:
A good example was the Jalozai camp, near Peshawar which was
established in 1980-81byAbdul Rasoul Sayyaf, the leader of the Ittihad-i
Islami (Islamic Unity) Party ... [and] became an important mujaheddin training
camp, with arms and ammunition depots; barracks and family quarters; a
medical college and hospital; schools and madrassas; and an office of
Pakistans Inter-Services Intelligence (ISI) (Turton and Marsden 2002, 11).
Not all refugee villages, however, were like this; others mainly functioned as safe havens
for the families of the mujahideen fighters.
Page 7 of 44
Afghan non-governmental organisations (NGO) provided another form of refugee

organisation for middle class urban refugees willing neither to join mujahideen factions
nor emigrate to the West and became an alternative way to channel assistance to the
refugee community, especially in the areas of health and education, sometimes
responding more to donor than to community needs (Azarbaijani-Moghaddam et al.
2002). As with the mujahideen parties, humanitarian assistance mixed with political
goals.
Countries of Asylum
Iran
Although Iran was a signatory to both the Refugee Convention and its subsequent
protocol, it chose to give Afghans the status of mohajerin, or people who seek exile for
religious reasons (Turton and Marsden 2002, 14). Many refugees were actually guest
workers whose status changed to that of refugees sur place as a result of the Soviet
invasion of Afghanistan. Furthermore arriving refugees were only housed for short
periods in transit camps before moving into wider society. Absorption into local labour
markets was facilitated by the sharing of the Persian language between the hosts and
the bulk of the refugees in Iran. While these factors may have contributed to Iran
receiving less attention than Pakistan, the most important reason for sidelining Iran may
have been its regional and international politics and standing. Shortly after the start of
the Afghan crisis, the 1979 Iranian revolution brought an Islamic radical regime to power,
and with it, Iran fell out of favour with Western countries, particularly the US (Benard and
Khalilzad 1984). Thus, even though Iran hosted nearly as many Afghan refugees as
Pakistan, and Afghan refugees in Iran comprised one of the worlds largest refugee
populations, it did so largely without international assistance (UNHCR 2000, 117). The
same year as the 1979 Iranian revolution, Iran also witnessed the US hostage crisis. The
resulting tensions in the relationship between Iran and the West probably explain why
Iran did not, at least initially, seek international assistance in dealing with the influx of
refugees (Turton and Marsden 2002, 11). Under pressure from Shiite refugees from
Iraq, Iran did request assistance from UNHCR in 1980 (UNHCR 2000). However,
International assistance to Iran was not forthcoming and UNHCR wrestled with the
disparity between the international response to the refugee crises in Pakistan and Iran
(UNHCR 2000). According to UNHCR (2000), Pakistan received in excess of US$1
billion over the course of twenty years (1979-1997) while Iran had to be content with a
mere US$150 million (UNHCR 2000, 118).
Wilkinson (1997, 15) observed that until now, Iran has been fiercely proud of its
ability to handle the refugee influx and consequently has limited both the international
and NGO presence in the country. However, impatient with the duration of displacement
due to lack of improvement in Afghanistan, from 1997 the Iranian government refused to
register new arrivals from Afghanistan, and the police have stepped up their random
questioning of Afghans in the street, considering most Afghans as illegal immigrants
(Turton and Marsden 2002, 15). Furthermore, under US pressure over Irans uranium
enrichment program, some circles in Iran may wish to stir up some trouble in
Page 8 of 44
Afghanistan in order to send a message to Washington of some of the risks of pushing

Iran too far.
These all may be contributing factors for Irans pursuing a policy of forced return.
since April 2007, the Iranian government has moved actively to expel Afghans who lack
formal papers permitting them to reside. Many are now housed in IDP camps near Herat,
and some reports suggest that their ranks comprise a significant number of ethnic
Pashtuns, whose anger at their precipitate expulsion from Iran has been a recruiting
boon for the Taliban movement in the hitherto relatively stable area of western
Afghanistan. According to UNHCR (2007, 257), Iran deported some 250,000 Afghans
last year. Even though the majority of deportees were single males, the humanitarian
situation of over 4,500 affected families made it a politically sensitive issue in
Afghanistan (UNHCR 2007c, 257).
Pakistan
Even though Pakistan is not a party to the 1951 Convention Relating to the Status of
Refugees or its 1967 Protocol, in recognition of its role as host for millions of refugees,
Pakistan is nonetheless a member of the Executive Committee of the Office of the
United Nations High Commissioner for Refugees.2 Although Pakistan proved notably
hospitable in a macrocosmic sense, accommodating refugee numbers on a scale that
puts to shame wealthier countries seeking to block the arrival of far smaller numbers of
forced migrants, at the microcosmic level, the refugees found themselves acutely
vulnerable to harassment by both Pakistani and Afghan forces. Refugee camps offered
a difficult environment where women faced constraints they had not had to endure in
Afghanistan (Boesen 1988, Dupree 1992, Centlivres-Demont 1993), where many petty
irritants blighted refugees daily lives (Saikal and Maley 1986), and where the milieu
offered easy access for those seeking to mobilise and utilise a socially-dislocated
generation of young men ripe for recruitment (Maley 1998).
Pakistans policy toward Afghan refugees needs to be seen in its historical and
geopolitical setting, both vis--vis India, but also Afghanistan itself, possibly linked to
Pakistans ambitious political quest for strategic depth in the region (see Rubin et al.
2001, 7). Initially, Afghan refugees were a means for Pakistan to become an important
player in the region as the middleman between international assistance and the
refugees. This helped legitimise the position of General Zia-ul-Haq, Pakistans ruler from
1977 to 1988, who had been marginalized due to his seizing power and hanging his
predecessor, Zulfikar Ali Bhutto (Grare 2003, 78). In essence, the more Afghan refugees
Pakistan hosted, the more generous was the assistance from outsiders, and the more
powerful Pakistan would become. Grare (2003, 68) labelled this behaviour calculated
kindness. By supporting refugee warrior communities (first the mujahideen parties and
later the Taliban), Pakistan could play a crucial role in determining who was to rule
Afghanistan. In order to counterbalance India's power, Pakistan has always tried to
create a friendly Islamic regime in Afghanistan. Pakistans move to control the refugee
2
http://www.unhcr.org/excom/40111aab4.html accessed 29 February 2008
Page 9 of 44
camps through the flow of humanitarian assistance, however, was also linked to a fear
that the refugees could become a geopolitical threat, mainly by awakening the old
Pashtunistan question and thus creating unrest in Pakistan while. At the same time
Pakistan could mend a previous breakdown in political relations with Afghanistan (Grare
2003, 65-66).
During the war against the Soviet-backed Afghan government there was an
overlap between the agendas of the Afghan parties and their Pakistani hosts, although
the Pakistanis tended to support religious radicals rather than more secularized
nationalists and royalists from within the refugee ranks (Hussain 2005). With the collapse
of the communist regime in April 1992, some parties shifted the focus of their activities
back to Afghanistan, while others saw their support bases erode. Nevertheless, refugee
populations have remained significant recruiting grounds for the radical Hezb-e Islami
(Islamic Party) led by Gulbuddin Hekmatyar, for the Taliban in the mid-1990s, and for
the current neo-Taliban movement. The Taliban was not a traditional force, but
emerged from a generation of refugees that neither knew their homeland nor peace
(Rashid 2000, 32), and, like the mujahideen parties before, also benefited directly from
the patronage of the Pakistans Inter-Services Intelligence directorate (ISI).
Despite its assistance to Afghan mujahideen parties, Pakistan was keen on calling
the refugee camps Afghan refugee villages (ARV) in order to avoid the connotation of
camps being linked to some form of military training. For the same reasons Pakistan
discouraged actual training of the mujahideen fighters in the camps so that it could deny
any involvement in the war (Grare 2003, 71). The relationship between Pakistan and the
Afghan insurgency was akin to holding a Tiger by its tail with the risk of getting savaged
when letting go. The recent displacements of Pakistanis into the eastern provinces of
Afghanistan due to insecurity in the Federally Administered Tribal Areas (FATA) are a
worrying step in this direction.
Even though the reception of Afghan refugees in Pakistan was initially very
generous, the honeymoon slowly came to an end in the 1990s when it became apparent
that there was no swift resolution in prospect for the refugee populations problems
(Schmeidl 2002). The growing discontent with Afghan refugees can be attributed to
several factors linked to the relative size of the refugee population, the duration of
displacement, and national and regional political developments. First, the sheer size and
length of displacement meant that many Afghan refugees outstayed the traditional
religious and tribal hospitality of Muslim countries (Centlivres and Centlivres-Demont
1988b, Rogers 1987). The longer refugees stayed, the greater the environmental impact
(in the form, for example, of deforestation around camps) and the higher the stress on
Pakistans infrastructure and resources such as schools, hospitals, land, and water
(Grare 2003, 63-64). According to Centlivres (1993, 33), the Pakistan government and
population held Afghan refugees accountable for ecological disturbances, rising prices,
spreading diseases and stress on the demographic balance.
Second, initial massive international assistance turned into donor fatigue shortly
after the Cold War incentive to finance the mujahideen resistance fighters vanished with
the withdrawal of Soviet forces from Afghanistan in 1989. Assistance to Afghan refugees
Page 10 of 44
began to decline in the 1990s and only picked up again at the end of 2001 when, with
the fall of the Taliban regime, massive repatriation was anticipated (Margesson 2007).
Turton and Marsden (2002, 15) consider the aid decline perhaps the most important
reasons for the change in attitude by the Government of Pakistan toward Afghan
refugees, as it meant a burden shifting from rich western states onto its own shoulders.
A third factor was resource competition between refugees and the host population
(over water, firewood, and employment). As Afghan refugees initially received
international assistance, many lowly-skilled Pakistani labourers were resentful that
Afghans benefited from the international humanitarian sector while seemingly draining
scarce Pakistani economic resources at the same time. This was especially felt in the
North West Frontier Province (NWFP) near the border between Afghanistan and
Pakistan, where Afghan refugees were concentrated.3
Later on, issues of insecurity were added to this long list, such as Peshawars and
other cities growing social ills, including crime, the widespread availability of weapons,
drug abuse, prostitution, and the decline in the Pakistan economy (Ruiz 2001, 4; see
also Grare 2003, Cheema 1988). Indeed, Afghan refugees did move frequently between
Pakistan and Afghanistan with Pashtuns in particular continuing traditional trade
networks, and the mujahideen resistance fighters and later also the Taliban constituting
a refugee warrior community that operated out of refugee camps. Furthermore, the war
in Afghanistan introduced massive quantities of small arms into Pakistan and provided
capital for investment in smuggling (Rubin et al. 2001, 9). More recently when the
Pakistan government increasingly came under international pressure over the growing
insurgency in Afghanistan, it was easy for the Pakistan government to make refugees
the scapegoats for Pakistans internal trouble and its own failure to curb the growth of
fundamentalism within its borders, as well as cover up for its own contribution to the
problem through its continuing to support the Taliban. While contemporary insurgency
fighters in Afghanistan have links to some refugee camps in Pakistan (such as the
Chaghi and Gerdi Zangal camps in Baluchistan), blaming an entire refugee population is
far-fetched, especially as camp closures so far have not lead to a reduction of instability
in Pakistan or insurgency activity in Afghanistan. Both have likely increased in the past
year despite camp closures. The December 2007 assassination of former Pakistan
Prime Minister Benazir Bhutto that plunged Pakistan into an internal turmoil unfortunately
contributed further to the negative trend of scapegoating refugees.
All this influenced Pakistans new stringent policy of voluntary return of Afghan
refugees, mainly via the forced closure of refugee camps, starting in 2004 with camps in
South Waziristan. The goal, which was not achieved, was to close all camps by late
3
Afghan refugees, who had initially lived mainly in the rural area of the NWFP, began to move into
the urban areas of Peshawar (creating an entire suburb - Hayatabad - of Afghan refugees), but
also Islamabad and the economic centre of Karachi in the south of Pakistan. This made Karachi
and Peshawar the cities with the largest Pashtun populations outside Afghanistan (Rubin 2000,
Ruiz 2001). While they boosted the economy by opening small businesses and becoming a
source of cheap labour (Ruiz 2001, 4), many Pakistanis blamed the refugees for driving up the
price of rental accommodation (Cheema 1988). In 1985, one in every 6 people in the NWFP was
an Afghan refugee (Centlivres and Centlivres-Demont 1988b, 72).
Page 11 of 44
2007. UNHCR appealed to the Pakistan government not to close refugee camps before
mid-2008 in order to avoid a humanitarian crisis in Afghanistan. This may harden
Pakistans negative policy toward Afghan refugees. However, the aftermath of the
February 2008 Pakistan elections may also possibly pre-occupy the Pakistan
government to the extent that the repatriation program may be less forcefully pursued.
Regardless of the roles Pakistans various regimes had in creating and supporting the
Taliban, the US chose to stand behind Pakistan policies, including those of scapegoating
refugees for allegedly contributing to insecurity and terrorism. Recent events in Pakistan,
however, have shaken US strategic thinking.
The United Nations High Commissioner for Refugees
UNHCR entered the picture after Pakistan called for assistance in 1979 (In Iran, UNHCR
was not permitted to work until 1983). It opened its first permanent office in Peshawar in
1980 (Grare 2003, 73) and five years later in Iran (Abbasi-Shavazi and Glazebrook
2006). As aid was administered through the Government of Pakistan, UNHCR was never
able to act in a fully independent manner. Similarly in Iran, UNHCRs access was initially
limited to specific settlements only. Nevertheless, despite a more limited role in Iran, the
UNHCR role was less politicised there than in Pakistan.
A major point of contention between UNHCR and the Government of Pakistan was
an early and ongoing disagreement over how to determine refugee status, as Pakistan
took registration through a mujahideen party as the basis for providing government
assistance (Schch 2008). This created ongoing tensions between UNHCR, Pakistan
and the Afghan mujahideen parties, with UNHCR generally losing out (Grare 2003, 74).
A second problem was the channelling of aid via middleman in refugee camps (see our
earlier discussion) handing control over camps to the resistance factions that, in effect,
militarised the refugee camps (Turton and Marsden 2002, 11). Loescher observed that
UNHCR and the World Food Program had no way of ensuring that food went exclusively
to non-combatants (Loescher 1993, 88-89). Donors looked the other way when the
Pakistani intelligence service operated in refugee camps and controlled international
assistance funds (see Zolberg et al. 1989; Rashid 2000; Turton and Marsden 2002).
This went as far as refugee villages boasting tanks and heavy artillery (UNHCR 2000,
248).
Terry (2002, 55) criticises the absence of debate concerning the ethical issues
or the implications for the safety of the refugees regarding such an acceptance by the
aid community that refugee camps were used by resistance fighters. Schch (2008)
argues that UNHCR was caught in a typical humanitarian dilemma forcing it to accept
infringement of its principles in order to continue its work and provide assistance to
refugees. It appears this may have set a precedent for how host (but also donor)
countries thought they could influence UNHCRs action and assistance in the region, as
currently UNHCR seems caught in a very similar dilemma to that which existed twenty
years ago having to compromise principles to continue receiving assistance for aiding
Afghan refugees. The only difference may be that, with the current repatriation program,
UNHCR found itself alone on the dance floor [and] unable to forge timely and effective
Page 12 of 44
partnerships with development organisations to ensure the successful reintegration of

refugees (Turton and Marsden 2002, 36; see also UNHCR 2006c):
Recognizing that tensions would eventually emerge over the scope and
duration of the agreements on voluntary repatriation, [UNHCR] launched a
policy initiative in mid-2003 to explore more comprehensive approaches.
While supporting voluntary return as the preferred durable solution, it argued
that a purely humanitarian and refugee-oriented perspective would be
insufficient to address the more complex challenges of development, poverty,
migration and demography that have emerged.
Afghanistan currently is the second largest operation for UNHCR after Chad, and
certainly the biggest in the Asia Pacific, even bigger than Iraq. In the global appeal for
2008-2009, the main focus is still on repatriation of refugees, but emphasis is also given
to dealing with the IDP problem inside Afghanistan. This is reflected in the distribution of
funds shown in Table 1, with more monies going to Afghanistan than to Iran and
Pakistan even though they continue to host the bulk of the remaining refugee
populations. This is despite UNHCRs recognising that the deterioration in the security
situation over the last two years, compounded by the increase in poppy cultivation,
threatens to overshadow the progress that has been made - although it does recognise
that maintaining the momentum of the voluntary repatriation programme and ensuring
sustainable reintegration will likely pose even greater challenges in the years to come
(UNHCR 2007a, 256). Already in 2004, UNHCR acknowledged that challenges in
Afghanistan are such that even an extension of existing arrangements for repatriation
sine die may resolve neither the immediate tensions between the rate of return and
absorption capacity inside Afghanistan, nor provide a definitive solution (UNHCR 2004a,
2). As such UNHCR considers that natural repatriation has come to a halt, with future
returns likely to respond mainly to push factors.4
Table 1. UNHCR Global Appeal 2008-2009: South-West Asia (USD)
2008
2009
Total
Afghanistan
$49,871,900 $50,595,410
$100,467,310
Pakistan
$18,374,173 $18,884,942
$37,259,115
Iran
$12,376,669 $12,273,117
$24,649,786
Regional
Total
$120,000
$412,962
$532,962
$80,742,742 $82,166,431
$162,909,173
Source: UNHCR Global Appeal 2007a
Email exchange with UNHCR official in Afghanistan, 27 January 2008
Page 13 of 44
There was early concern that large numbers of returns to a situation in which
these returns cannot be sustained will be detrimental both to the safety and human
rights of returnees as well as to the long-term reconstruction of Afghanistan (Amnesty
International 2003, 6). Still, donor fatigue and host country pressure, coupled with
seeing large-scale repatriation as a way of boosting the legitimacy of a fledgling state
apparatus, struggling to exert its authority in a situation where power is still contested,
may have forced UNHCR in 20002 to embrace return as a preferred durable solution for
Afghan refugees despite a recognition that repatriation aid was in the best interests
neither of the majority of its intended beneficiaries nor of the long term reconstruction of
Afghanistan (Turton and Marsden 2002, 35, 56). UNHCR let its concerns be known by
only facilitating and not promoting return to Afghanistan (Amnesty International 2003).
Yet such a distinction may be difficult to uphold in a context where both host countries
(Iran and Pakistan) forcefully push for refugee return and remaining refugees having no
strong desire to return home.
Even before the assassination of Benazir Bhutto, UNHCR was struggling for
humanitarian space in Afghanistan, and was not optimistic that it would have much
political space to manoeuvre in Pakistan for the next 3-5 years.5 With the assassination,
this period of uncertainty may prove to be even longer. One could argue that UNHCR is
merely trying to sit out a difficult political situation, especially in Pakistan. On the one
hand, UNHCR is focusing its advocacy on opposing Pakistans policy orientation on
principled protection and practical grounds, while on the other hand looking for positive
opportunities on which it can build, hoping that powerful international actors will pick up
and support its concerns.6 A UNHCR official in Afghanistan described their current work
mainly as reactive, struggling to put out one fire after the other.7
This, however, may force UNHCR to go along with the political games played by
both Iran and Pakistan, possibly be hostage to them, and become an accomplice to a
repatriation scheme that is questionable from a protection stance, as well as from the
perspective of sustainability and stability in Afghanistan. In 2004, Pakistans forced
closure of refugee camps in South Waziristan caught UNHCR off balance,8 resulting in a
situation where expulsion was accepted as voluntary return in order that UNHCR could
assist the refugees.9 As a pragmatic approach seems to prevail, UNHCR may already
have compromised its principles of protection by putting regional politics ahead of the
rights and needs of refugees (something acknowledged by UNHCR officials on the
ground).
Interview with UNHCR official, Kabul, Afghanistan, 11 November 2007

Interview with UNHCR official, Kabul, Afghanistan, 11 November 2007 and follow-up
commentary via email 16 February 2008
7
Interview with UNHCR official, Kabul, Afghanistan, 11 November 2007
8
Information provided on email by UNHCR offical in Afghanistan, 16 February 2008
9
Discussions with UNHCR officials in 2004 when Pakistan closed refugee camps in South
Waziristan and forced the residents over the border into the province of Paktika. While UNHCR
assumed that return was forced, it went along with the official Pakistani policy that return was
voluntary in order to be able to assist the refugees.
6
Page 14 of 44
All this clearly highlights the difficulty faced by an operational organisation that has
to stand for the rights of refuges while still having to function effectively on the ground
(see Maley 2003). One has to understand that UNHCR, like the UN in general, is largely
an organisation at the very least influenced by the wishes and priorities of states,
particularly donor and host states. Earlier in the history of the Afghan refugee crisis, it
struggled to obtain funds for those displaced to Iran and encountered political obstacles.
UNHCR once again seems caught in a difficult position where especially the US seems
to be putting pressure on the organisation to go along with Pakistans repatriation
program.10 This said, Turton and Marsden (2002, 56) find that UNHCR may have more
leverage than it appears to think, especially with the growing humanitarianism and focus
on responsibility to protect among western states. Furthermore, Loescher (2001, 350)
has argued that while state interests have shaped UNHCR as it is at the mercy of donor
funds, it has never been fully a passive mechanism of states.
The Current Situation of Refugees and IDPs
Overview of the Refugee population
According to the UNHCR Global Appeal for 2008-2009, there are currently still three
million Afghans refugees in exile, with two-thirds residing in Pakistan (2.1 million) and
one-third in the Islamic Republic of Iran (915,000) (UNHCR 2007a, 257). This statistic
already exposes the difficulty of sustainable return to Afghanistan. Despite over 5
million reported or recorded returnees, the remaining refugee population seemed to
remain stable, if not actually increasing. Retrospectively this led UNHCR to adjust the
initial refugee population to 8 million, about 2 million higher than estimates in the past.
While some ostensible growth of a refugee population can be attributed to fertility and
an inclusion of previously unregistered refugees (Kronenfeld 2008), a greater proportion
was linked to the so-called recycling problem return and re-entrance into Pakistan or
Iran in order to benefit more than once from return assistance packages (Turton and
Marsden 2002, 20).11 The practice of recycling by refugees here may seem less
surprising as it occurred during previous repatriation schemes (Centlivres 1993) that
UNHCRs readjusting their figures to such a high level, accepting refugee figures at face
value rather than questioning them.
Afghan refugee populations outside Iran and Pakistan are much smaller, with over
half residing in Europe (mainly in Germany, Netherlands, the United Kingdom and
Denmark, but also smaller countries such as Switzerland and Austria) and the other half
in Australia and New Zealand, North America (US and Canada), Russia, Central Asia,
and India. According to Orywal (1993-94), for example, at the end of 1992 there were
already more than 40,000 Afghans living in Germany. Early migration was largely limited
10
Interview with UNHCR official, Kabul, Afghanistan, 11 November 2007. It is important to note
the strong alliance between Pakistan and the US.
11
In a recent registration process of Afghan refugees in Pakistan, however, only 5.4 percent of all
remaining refugees admitted to having gone to Afghanistan and returned (Government of
Pakistan 2007, 14).
Page 15 of 44
to the wealthier and more highly educated urban population; many who had previous
links to western countries through having studied there (Boesen 1988, Centlivres and
Centlivres-Demont 2000). Later on, family reunification schemes, but also smuggling and
individual asylum applications, and to a lesser degree resettlement (see later
discussion), completed the picture. Labour migrants are also mixed in with the refugees,
showing the complexity of the situation. Nowadays, there are many Afghan migrant
workers to be found in the Gulf States. Statistics are difficult to interpret, as Afghans in
these countries are a mix of those who have been granted asylum and/or citizenship,
and asylum seekers. According to UNHCR (2005b) since 1994, when the Taliban first
emerged, some 238,000 Afghans have sought asylum in industrialized countries outside
the region. Germany, has received 50,000 since 1994. The next largest recipient is the
Netherlands, which has received 36,000 in the same period, followed by the U.K. with
34,000, Austria with 31,500, Hungary with 13,500, and Denmark with 11,500. The
current figure for Afghan asylum seekers with pending status claims is 14,974 (UNHCR
2006b).12
Afghan Refugees in Pakistan
As noted earlier, Pakistan is not a party to the 1951 Refugee Convention (or
related 1967 Protocol), and thus the 1946 Foreigners Act (amended 2000) remained the
only law applying to refugees and asylum seekers even though it recognized no
distinction between them and other foreigners.13 This provided Pakistan with some
leverage in dealing with the refugees on its own terms, rather than having to adhere to
international standards. Only in January 2007 did the Pakistani government register
refugees in such a way as finally to give them a legal status for residing in Pakistan
(UNHCR 2007a, 257).14
Initially, the bulk of Afghan refugees in Pakistan were housed in what were first
called Refugee Tented Villages (RTVs) and then so-called Afghan Refugee Villages
(ARVs) as with the passage of time they acquired a more permanent appearance and
mud-brick structures replaced tents. Unlike other refugee situations, UNHCR was able to
set up 345 official ARVs where the majority of refugees lived (Centlivres 1993, 14;
UNHCR 2000, 116). According to Colville (1997), these refugee villages were not closed
off from the outside world and Afghan refugees were allowed to move freely ... [in order]
12
In addition to Europe, North America (the US and Canada) hosts about 27,000 Afghan
refugees, with over half residing in Canada (UNHCR 2007b, i). Australia hosts some 7,000
Afghan refugees. The figure in India is disputed, and lies between 8-30,000 Afghan refugees,
mainly of Hindu/Sikh origin. There are about 6,000 left in Central Asia (mainly Uzbekistan,
Tajikistan, Turkmenistan and Kyrgyzstan), with a majority having served in the former Communist
government or having fled more recent sectarian violence, therefore being unable to return (see
IRIN (2003), Turkmenistan: Afghan refugees want third-country resettlement (30 July 2003);
http://www.irinnews.org/report.aspx?reportid=20566, accessed 4 January 2008 and IRIN (2004),
Tajikistan:
Focus
on
Afghan
refugees
(16
February
2004),
http://www.irinnews.org/report.aspx?reportid=40399, accessed 4 January 2008.
13
http://www.refugees.org/countryreports.aspx?id=2013 [accessed 24 January 2008]
14
The registration exercise resulted in the issuance of Proof of Registration (PoR)
documentation valid until 2009 to 2.15 million Afghans (UNHCR 2007a, 257).
Page 16 of 44
to find work and set up businesses (Colville 1997, 4). Some refugees, often those
supported by funds from relatives abroad, moved to urban centres, especially Peshawar
city, Rawalpindi, and Karachi, and prior to repatriation efforts, refugees living outside
camps made up a majority of the remaining refugee population: 1.7 million out of 3.05
million (Collective for Social Science Research 2006, 3), many of them having acquired
Pakistani IDP cards. By the time of the registration exercise of the Government of
Pakistan in 2007, this figure had moved closer together for the remaining population,
with 55% of all refugees living outside camps (1.2 million) and 45% inside camps
(976,605), suggesting a higher trend of repatriation among the non-camp population. 15
Registered refugees in camps could access basic support supplied through the
Commissionerate of Afghan Refugees of the Northwest Frontier Province, Balochistan,
and Islamabad. While reports surfaced in the 1980s of multiple registrations that could
have artificially inflated refugee numbers, to some extent this problem was offset by a
countervailing phenomenon, namely the presence of unregistered refugees, including
urban dwellers (see Farr 1985, 104-105; Dupree 1987; 1988). As a UNHCR (2007b, i)
source put it, Before March 2005, ascertaining the number of Afghan refugees living in
Pakistan was always at best a calculated guess. In early 2005 (25 Feb-11 March), three
years after massive repatriation to Afghanistan had begun, the Pakistan government,
assisted by UNHCR, finally conducted the first ever census of Afghans who had arrived
in Pakistan over the previous 25 years, leading to a subsequent registration process.16
This prompted UNHCR to adjust its refugee figures from previous years as the census
came up with 3 million Afghan refugees residing in Pakistan. As UNHCR figures were
just shy of 1 million in 2004, an adjustment of close to 2 million had to be made (see
UNHCR 2006a, 8).17
Information from the registration process reveals that the remaining Afghan
refugee population in Pakistan mainly consists of young individuals whose families
arrived in the country at the onset of the conflict in Afghanistan, with 77% having arrived
by the end of 1988 and 50% in 1979-1980 alone (Government of Pakistan 2007, 13). A
majority of the Afghan refugees who now remain in Pakistan were either born in exile
(with 13% being under the age of five, and 55% 17 years or younger) or spent most of
their lives there, thus having no experience of living in their home country. This is also
15
The majority of Afghan refugees in Pakistan live in two provinces bordering Afghanistan where
the refugee camps are also located: the North West Frontier Province (NWFP, 64%) and
Balochistan (Quetta, 21%). The rest lives in the urban centres of Punjab (Islamabad, 9%) and
Sindh (Karachi, 4%). Currently, Afghans constitute six percent of NWFPs population, 5.9 percent
in Balochistan and 3.4% in Islamabad (see Government of Pakistan 2007).
16
Participation was mandatory for all Afghans who arrived in Pakistan after 1 December 1979,
the month in which Soviet troops invaded Afghanistan and triggered a mass exodus of refugees
to Pakistan and Iran. Only those who were included in the census would be eligible to be included
in a registration, which was to provide some form of identification.
17
Earlier estimates never accounted for the increase of from births and never captured out-of
camp populations living in cities, or the recycling of returnees (see Kronenfeld 2008). The U.S.
Committee for Refugees estimates that there are 2.4 million Afghan refugees in Pakistan, even
though
only
2.15
were
given
registration
cards.
Page 17 of 44
reflected that about two-thirds of all refugees are single (66%) and only 33% are married
(Government of Pakistan 2007, 11). Out of 46.8% women refugees, 4% are single heads
of households while 96% are considered female dependents. Overall there are more
male than female refugees across all age brackets (Government of Pakistan 2007, 8).
The education level of remaining Afghan refugees is also very low with 71%
reporting no formal education (Government of Pakistan 2007, 8). Only 20 percent of all
refugees participate actively in the labour market, with 71% having no effective monthly
income, and 89% reporting no skills (Government of Pakistan 2007, 8). Over half of all
refugee households obtained their livelihoods from daily labour. Lack of skills makes
return difficult and reduces the refugees prospects of finding an adequate livelihood
upon return (UNHCR 2004a). Because Afghan refugees in Pakistan came
disproportionately from provinces in the south and east of the country, Pashtuns made
up an overwhelming majority of the refugee population, well over 80% (Sliwinski 1989b)
and 93% of the camp population (Government of Pakistan 2007, see Figure 2).
Figure 2. Documented Afghans in Pakistan by Ethnicity (2007)
Pashtun
Tajik
Hazara
Uzbek
Turkme
n
Other
Source: Government of Pakistan (2007)
Afghan Refugees in Iran
Afghans who migrated to Iran and settled there during the 19th and early 20th centuries
were naturalised as Iranian citizens and came to be classified as an ethnic group known
as Khawari or Barbari (Adelkhah and Olszewska 2007, 140). Later on Afghan migrant
workers came to Iran after the 1960s, particularly during the terrible famine of 19711972 (Adelkhah and Olszewska 2007, 141). While Iran, as noted earlier, did ratify the
1951 Refugee Convention and the 1967 Protocol, and while Iran's 1963 Regulations
Relating to Refugees (1963 Regulations) provided that Refugees should not be forcibly
returned to the country where their life or freedom is endangered for political, racial or
Page 18 of 44
religious reasons or for their membership in a particular social group, Iran maintained
some reservations including the provision in the 1979 Constitution that disallowed
granting asylum to applicants if they are regarded as traitors and saboteurs. 18 In 1991,
when the Afghan refugee population in Iran peaked, the government finally decided to
grant refugees prima faci status issuing them blue cards (Adelkhah and Olszewska
2007, 141). This, however, did not mean that Iran gave refugees many rights19 (see also
Centlivres 1989, 12 regarding labour restrictions of Afghan refugees).
A majority of the refugees lived in big cities or surrounding rural areas (Adelkhah
and Olszewska 2007), although some resided along the border to western Afghanistan
(Turton and Marsden 2002). According to Wilkinson (1997, 14) this absorption of the
refugees into local communities [is] in sharp contrast to other regions, but also puts the
burden for finding a place to stay on refugees, who tended to concentrate in urban areas
in enclaves where other Afghans live (Abbasi-Shavazi and Glazebrook 2006).
The lack of refugee camps and a refusal to register refugees who arrived after
1997 (Turton and Marsden 2002) makes the counting of refugees very difficult:
according to Abbasi-Shavazi and Glazebrook (2006, 2), there are no official published
statistics on the substantial number of Afghans in Iran who are neither registered nor
counted as labour migrants. A recent (late 2005/early 2006) government registration
exercise (Amayesh-II), aimed to ascertain the number of Afghans living in Iran, put the
number of officially registered Afghan refugees in Iran at 920,000 (UNHCR 2007b, i).
This, however, only included re-registered refugees who had arrived and registered prior
to 2001, not counting new arrivals or those that never had been registered (Margesson
2007). UNHCR estimates another half million Afghans among undocumented refugees
and labour migrants (Abbasi-Shavazi and Glazebrook 2006, 2).
Figure 3 provides an overview of Afghans in Iran by ethnicity, showing that, in
contrast to Pakistan, there are very few Pashtun refugees while the Hazara (a minority in
Afghanistan) and Persian speakers (Tajiks) make up the bulk (47 and 30 percent
respectively, Schetter 2003, 386).20 This reflects the kinds of cultural-geographical
pools from which the refugee population was drawn. In contrast to Afghan refugees in
Pakistan who came from the Pashtun-dominated South, East or Southeast those in Iran
originated largely from the Northern (26%), Central (24%) and Western provinces (13%)
18

According to Iranian law, property ownership of foreigners is illegal, and such ownership by
Afghans is uncommon. The exception to this is Mashad, where a system of informal property
ownership has evolved that allows Afghans to informally purchase houses (Abbasi-Shavazi and
Glazebrook 2006, 4). Furthermore, Afghans are restricted to certain types of manual labour in
Iran (Abbasi-Shavazi and Glazebrook 2006, 7, see also Adelkhah and Olszewska 2007, 143).
However, they were still able to find work, not least because so many Iranian men were
conscripted to fight in the war against Iraq which began in September 1980 (UNHCR 2000, 116).
20
The Ameyesh-II exercise shows that half of all Afghan refugees in Iran are from the Northern
provinces of Samangan, Faryab, Balkh and Badakhshan (26%) and Central provinces of Parwan,
Kabul and Ghazni (24%) (UNHCR 2007b). The remainder are from the Western Region (13%,
Herat and Farah) and South (8%, Kandahar, Uruzgan), with another 8% having fled the Central
Highland region, which is the traditional territory for the Hazara population. Refugees from the
East and Southeast are in the minority (2%).
19
Page 19 of 44
(UNHCR 2007b). The proportion of Hazaras in the ethnic breakdown had risen by 6
percent since the beginning of 2004, suggesting that they are the ethnic group that is
most reluctant to repatriate (Adelkhah and Olszewska 2007, 143).
Figure 3. Documented Afghans in Iran by Ethnicity (2005)
Hazara
Tajik
Pashtun
Baluch
Uzbek
Turkmen and other
Source: AMAYESH-II registration exercise (Iran 2005/6)

In contrast to Pakistan, less than half (40%) arrived by 1985, with 36% having
entered Iran after 1996 (Adelkhah and Olszewska 2007, 143). There are more men
(56%) than women (44%) corresponding with earlier figures that put the percentage of
adult men among Afghan refugees in Iran at over 60% in the 1980s (Centlivres 1993).
The Amayesh statistics indicate that the vast majority of registered refugees were living
as families (89 percent), with a mean family size of 4.7 persons (Adelkhah and
Olszewska 2007, 143), which stands in contrast to the single male domination of
unregistered refugees and migrants. As in Pakistan, political parties had an influence
among the Afghan refugees in Iran, mainly the Hezb-e Wahdat (Unity Party) targeting
mostly Shia refugees (such as the Hazara). Education levels among Afghans in Iran
seem higher than those in Pakistan (see Adelkhah and Olszewska 2007) with 90 percent
of all refugee families in Mashad being literate (Abbasi-Shavazi et al. 2005).
As noted earlier, Iran has pursued a strong policy of pushing Afghan refugees back
home since the fall of the Taliban in late 2001. This has been done through various
measures, such as restricting employment and freedom of movement, levying taxes,
launching mass roundups of unregistered Afghans, and revoking the refugee cards of
Afghans arrested for petty offences.21 While in the past Afghans had equal access to
education and health care, they are now forced to pay the same rates as other
foreigners, and the Iranian government closed informal schools used by Afghan
21
http://www.refugees.org/countryreports.aspx?id=2001
[accessed
24
January
2008].
Furthermore, Iranian women who marry Afghan men lose their Iranian citizenship (Strand et al.
2004, 3).
Page 20 of 44
refugees (Monsutti et al. 2006). Worst of all, however, is physical deportation (nearly one
million over the past five years (UNHCR 2007a)), including some 250,000 illegal Afghan
migrants; mainly single males (UNHCR 2007c, 257). This has made life in Iran difficult
for Afghans, who had few rights to begin with and have been exploited for cheap labour
(sometimes not receiving proper wages) and face high rental prices for property. That
said, labour migration between Afghanistan and Iran, clandestine or not, was and still is
very active, because job opportunities and salaries are nevertheless higher than in
Pakistan.
Overview of Internal Displacement
In its Global Appeal for 2008/9, UNHCR defined assistance to internally displaced
persons (IDPs) in Afghanistan as one of its major priorities. Internal displacement has
been part of the Afghan wars as much as refugee movement, but, even more than in
other parts of the world, has been poorly documented. Since 2002, UNHCR has assisted
about half a million IDPs to return home (UNHCR 2007b, iv).22
According to UNHCR, at the end of 2007, about 130,000 individuals were still
displaced within Afghanistan, mainly (86%) in the South living in camps (particularly in
the Southern region [mainly Kandahar and Helmand] in Maiwand, Mukhtar, Zahir Dasht
and Panjwayi Camps) and camp-like situations [where they] are in need of assistance
and/or protection (UNHCR 2007b, iii). The Afghan Red Crescent Society (ARCS)
recently described conflict-induced displacement in Southern Afghanistan as a major
22
There are various reasons why documenting IDPs in Afghanistan is difficult:

Due to frequent displacement in Afghanistan, IDPs may not consider smaller movements
over short distances as worth acknowledging, especially if displacement is of short
duration and accommodation is with relatives.
IDP figures often fluctuate greatly as it is not uncommon that families flee during fighting,
but return during the cultivation season to tend to their land or poppy harvest or to work
and tend the fields. For example, persons displaced during fighting between the Taliban
and NATO in Musa Qala, Helmand in 2006 returned home after violence ceased.
Similarly, while the battle of Chora in Uruzgan (15-19 June 2007) created extensive
displacement in the area, many were able to return to their villages a few months later.
Humanitarian access to the area with the majority of internal displacement (the South) is
currently extremely limited. There are only a few IDP camps in existence, such as in the
capitals of the three Southern provinces of Helmand (Lashkar Gah), Uruzgan (Tirin Kot),
and Kandahar. Often IDPs live in poor conditions (with respect to health, sanitation,
nutrition, education) due to lack of assistance. Lack of assistance also forces many IDPs
to settle with relatives, with the more affluent ones renting houses, thus falling outside
most estimations.
Fourth, prospects of assistance to IDPs may have also contributed to exaggerated
figures in Helmand. For example, while a local Afghan NGO working with IDPs in Lashkar
Gah estimates that there are about 3,000 IDP families from Nawzad, Musa Qala,
Garamsir and Sangin districts, government officials have raised the number to 7,000
families in order to receive refugee ID cards (and associated assistance and benefits) for
their families and tribes. A local NGO reports that a Ministry of Rural Reconstruction and
Development driver was caught with 50 refugee cards that he wanted to distribute to
friends and family, causing WFP to refuse to accept new applications, which in the end
harmed IDPs with legitimate claims.
Page 21 of 44
humanitarian challenge (IRIN, 20 November 2007).23 The Internal Displacement

Monitoring Centre (2007), estimates IDPs in Afghanistan to be around 300,000,
including the following groups:
IDPs due to ethnic oppression, such as Pashtuns from the north who were forced to
leave due to systematic persecution, though not all are in the caseload of UNHCR.
There are 3,285 IDPs in the Northern region (Gujur Afghans, an ethnic minority in the
North of Afghanistan, originally from Takhar) who cannot return home due to
protection problems (Rahjo 2007). Far less documented are those displaced by tribal
(or factional) disputes in the South and elsewhere.
Recent and ongoing displacements from insecurity and escalation of conflict in the
South (insurgency) and East. Violence and fighting between NATO and Afghan
National Army forces and the insurgency had displaced about 20,000 families (about
100,000 individuals) in the provinces of Helmand, Kandahar and Uruzgan by the end
of 2006.24 This figure increased drastically in 2007 when the UN estimated in August
2007 that some 80,000 people have been displaced by insecurity, predominantly in
the south, southwest and east of the country (IRIN, 16 August 2007)25 and
displacement is continuing.26 UNHCR has raised its concern with both the Afghan
government and international military forces about the protection of these individuals.
They are generally unassisted due to lack of access. Two groups can be
differentiated:
o Political IDPs consisting of families related to pro-government strongmen who
flee out of fear of persecution from the Taliban insurgency in their native
district.
o Those who flee more generally from existing or anticipated violence and
fighting in districts of Taliban insurgency. Many flee proactively, fearing
NATO/US bombardment in areas where Taliban has taken control; many do
not return out of the same fears. 27
Returning refugees/migrants unable (or unwilling) to return back to their
province/district of origin such as deportees from Iran having to settle in IDP camps
in Herat.28 This may be due to a general lack of the rule of law, widespread
23
Afghanistan: Conflict-affected displacement major humanitarian challenge - Afghan Red

Crescent, 20 November 2007 (IRIN) http://www.irinnews.org/Report.aspx?ReportId=75399,
accessed 1 March 2008
24
http://www.internal-displacement.org/
8025708F004CE90B/(httpCountries)/DFADB5842F9262BF802570A7004BA6F0?OpenDocument
25
Afghanistan: UN highlights conflicts impact on civilians,16 August 2007 (IRIN)
http://www.irinnews.org/Report.aspx?ReportId=73759, accessed 1 March 2008
26
Afghanistan: Hundreds flee fighting in Helmand Province, 06 December 2007 (IRIN)
http://www.irinnews.org/Report.aspx?ReportId=75722, accessed 1 March 2008
27
For example, insurgency activities and international bombardments of insurgents as well as
heavy fighting in certain areas have caused internal displacement in Uruzgan, mainly from Chora,
Deh Rawud, Shahidi Hassas and the Taliban stronghold areas of Mehrabad and Darafshan in
Tirin Kot. In Helmand this includes displacement in Musa Qala, Sangin, Girishk, and Nawzad.
28
Afghanistan-Iran: Afghan deportees complain of lack of aid , 14 August 2007 (IRIN),
http://irinnews.org:80/Report.aspx?ReportId=73721, accessed 1 March 2008
Page 22 of 44
landmines and a dearth of livelihoods (see Afghanistan Independent Human Rights

Commission 2007). This also includes urban IDPs in Kabul - mainly refugees from
other areas in Afghanistan who have returned to the capital for lack of employment,
shelter or economic survival elsewhere (Koser 2007). UNHCR has knowledge of
3,600 IDPs in the Kabul region, but there could be many more considering that the
population of Kabul has grown drastically since 2001 (Rahjo 2007).
IDPs who have been forced to move due to drought and lack of economic
opportunities in their native provinces or districts. Food security is a major issue in
Afghanistan, especially in the West (for example Ghor), which is often exacerbated
during hard winter months. Pressure on poor rural areas by returning refugees
creates a situation of competition over scarce resources.
There is also a new form of development-induced displacement affecting mainly
poor urban dwellers living on borrowed land. The most publicised displacement was
the forced displacement of urban dwellers from a squatter settlement close to one of
Kabuls most prominent neighbourhoods (Sherpur) after the Ministry of Defence reclaimed its land to distribute it among politicians. Other displacement is caused by
the rising prices of land in urban areas and illegal land-grabbing, again mainly
affecting poor urban dwellers.
The hardest form of displacement to assess might be that of Kuchis as they are
nomadic people constantly on the move. More recently, however, ethnic tensions (for
example in Wardak), increased poppy cultivation, and illegal land seizures have
infringed upon their traditional grazing land forcing them to go elsewhere. There are
no figures available on this problem.
All-inclusive solutions for IDPs in Afghanistan do not exist at present: There is lack
of a comprehensive strategy with different instruments in place that can meet the needs
of IDPs.29 The two entities working on assistance are the Directorate for Refugees of the
Afghan government, and UNHCR, with the former lacking resources and the latter
lacking humanitarian access, mainly working through local non-government
organizations or the Afghan government.30 Governmental assistance to IDPs overall has
been very poor. A local Afghan NGO has alleged, for example, that IDPs in Helmand
were exploited as many of the government departments distributed food ration cards
meant for the displaced population among the their own staff and clientele.31 A recent
29
Walter Kalin, UN Representative for IDPs as cited in IRIN Afghanistan: UN highlights conflicts
impact on civilians,16 August 2007 (IRIN) http://www.irinnews.org/Report.aspx?ReportId=73759,
accessed 1 March 2008
30
Even in traditional areas known for assistance to IDPs, things have changed. Many IDPs, for
example, moved to Helmand as it is known as a province where the Government has traditionally
distributed land to landless communities since the 1950s. The government of Helmand, however,
has not taken any initiative in regard to the current displaced communities in the province. The
Director of the Refugee/Immigration Department argued that the former Governor, Engineer
Daoud, did not want to allocate a specific location for displaced families because this would
encourage other communities to move to the IDP camp in Lashkar Gah.
31
Interview with local Afghan NGO. Furthermore, teacher salaries were less than originally
assigned by the Ministry of Education.
Page 23 of 44
policy discussion centred on the use of IDP camps, as there is hope that people will
return faster when living in non-camp environments.32
Protection Challenges
About 14 percent of all refugees in Pakistan cited special needs during the registration
process conducted by the Government of Pakistan with 58% citing legal and physical
protection needs with special medical conditions also being noted (AREU 2007, 3) .
Furthermore, female-headed households, children and youth are added to the risk
categories (AREU 2007, 3). More crucial, however, is the current situation in
Afghanistan, which makes for a very difficult protection environment. There has been
overall a lack of concerted reconstruction efforts, keeping Afghanistan at the bottom of
the Human Development Index.33 A great number of landmines remain uncleared;
farmers lacking other livelihoods are turning increasingly to poppy production. Returnees
to Afghanistan have to cope with insufficient services in the areas of health care,
education, and basic infrastructure. This makes it especially difficult for women, children,
elderly, sick and handicapped individuals. The new Afghan state is in no position to offer
short-term solutions to these problems: the substantial collapse of the Afghan state
during years of war means that, to this day, much of the activity of the new post-Taliban
state is funded from external sources, with Afghans consequently subjected to the policy
priorities of donors (Maley 2006).
According to a recent report by the Afghanistan Independent Human Rights
Commission (AIHRC 2007)34 on economic and social rights in Afghanistan, based on
interviews with over 11,000 people (focussing, among others on returnees), the following
key challenges exist for those living in and returning to Afghanistan:
Livelihood: 62.3% of all interviewees reported no stable income within their
households, with more than 60% living below the poverty line, earning less than
US$1 per day. UNHCR (2004a, 15) also identified landlessness or access to
agricultural land as a key challenge, as about 70% of returnees claimed to have no
land. This is a especially a problem for the Kuchis, the Afghan nomadic pastoralists,
whose access to grazing lands has been diminishing due to drought and conflict
(UNHCR 2004a).
Shelter: 67.1% of interviewed returnees chose not to return to their places of origin
due to lack of shelter; and 67.3% left their places of origin for the same reason after
returning there. Similarly 43.4% of IDPs claimed lack of shelter as a reason for
displacement. There are also problems with supportive infrastructure, such as
potable water, sanitation, roads, transportation and irrigation canals that impact on
livelihood (UNHCR 2004a).
32
Afghanistan: Should conflict-displaced IDPs go into camps? 28 November 2007 (IRIN Radio)
http://www.irinnews.org/Report.aspx?ReportId=75542, acessed 1 March 2008
33
Abouth one fourth of all Afghans live at the fringes of poverty; in some provinces this is as high
as 50-90%.
34
The research, based on field monitoring activity, was conducted over the period of year
(January-December 2006) in 32 of Afghanistans 35 provinces in partnership with UNHCR.
Page 24 of 44
Adequate health care: Although over 80% had access to governmental/NGO health
care services, about 40% either had no access or felt the service provided and staff
skills were of poor quality.
Access to (primary) education: A third of all interviewees stated that their primaryschool age children (mainly girls) did not attend school regularly, either because of
lack of school buildings (girls) or the need for child labour (boys).
The human rights context also needs careful consideration. Many rights only
exist on paper, especially for women who are still subject to discrimination at the hands
of the Afghan justice system, especially traditional conflict resolution mechanisms. A
majority of all Afghans still rely on traditional justice mechanisms (AIHRC 2007), which
are considered a common source of violation of rights (for example, lack of due
process) (Stapleton 2007b, 19).35
The last two years have seen an increase in insurgent activity and violent incidents
in Afghanistan. UNHCR (2004a, 14) also raised a concern about the ethnicisation of the
Afghan conflict and the presence of controversial figures in Afghanistans ongoing
political process. The year 2007 witnessed 20-30% more attacks than in 2006 and in
2008 another increase is expected (Oxfam 2007). Intensified warfare by the insurgents
has been met very forcefully by the ANA and ISAF/NATO, raising many concerns about
civilian casualties during aerial bombing. According to an Oxfam report (2007), there are
four times as many air-strikes by international forces in Afghanistan as in Iraq (Oxfam
2007, 16).
The strategy of the insurgency builds on asymmetric and psychological warfare,36
while at the same time exploiting the mistakes made by the Afghan National Army (ANA)
and its international supporters. The insurgents approach that if you are not with us,
then you are against us make it difficult for anybody to stay neutral. Teachers in
particular, but also health workers, and recently anybody seen as associated with
international forces and what are considered pro-government clergy, are targeted.
Added to this are systematic acts of intimidation by regional warlords, militia
commanders, criminal groups and narcotics dealers; many with links to the Afghan
government. This includes illegal land occupation, which is fuelling land disputes all over
Afghanistan. Competition over resources (mainly land and water), but also ruthless
government officials abusing their power, has caused an increase in tribal conflicts,
including blood feuds. The increasing war economy has also resulted in entire areas
being under the control of drug mafias and smuggling networks, which tend to work in a
mutually beneficial relationship with the insurgency.
35
There is no justice for converts and homosexuals. For example, a number of successful asylum
claims by Afghans in Australia have been based on the feared consequences of either conversion
or homosexuality. While freedom of speech is guaranteed and Afghanistan has one of the most
progressive media laws in the region, there are questions about the scope of this freedom, with
media coming frequently under attack or practising self-censorship.
36
This section draws from unpublished research in the south of Afghanistan by a local Afghan
NGO during 2007.
Page 25 of 44
As a result, the local population, especially in the volatile South, but increasingly
also in the East and Southeast and elsewhere, are caught between a rock and hard
place. On the one hand they have to deal with a malfunctioning, and often highly corrupt,
Afghan government that is unable to provide basic security, protection and even a
semblance of the rule of law. On the other hand, they are confronted by a growing
insurgency that is supported internationally by Pakistan and Al Qaeda. The latter has led
to cross-border infiltrations of foreign Taliban into Afghanistan (such as Urdu, Punjabi
and Arabic-speaking militants). As security has become a scarce commodity,
communities increasingly accommodate the insurgency.
Last but not least, the lack of access for UNHCR and other international agencies
in many parts of Afghanistan makes protection, and especially the monitoring of
returnees, difficult (Amnesty International 2003, 28) and the space for humanitarian
action is diminishing every day, especially as the insurgency has made the kidnapping of
aid workers part of their psychological warfare. In addition to this overall difficult
protection environment in Afghanistan, UNHCR (2006d, 1) divides extremely vulnerable
cases into two broad categories:
(i) Individuals whose vulnerability is the result of a lack of effectively
functioning family- and/or community support mechanisms and who cannot
cope, in the absence of such structures.
(ii) Individuals who cannot cope, either because such support structures are
not available or because Afghanistan lacks the necessary public support
mechanisms and treatment opportunities.
This includes the following population groups (UNHCR 2006d, Rahjo 2007, Stapleton
2007b):
Deportees: Deportations from the Islamic Republic of Iran and the return of
unregistered Afghans from Pakistan highlighted the need to address population
movements outside the refugee framework (UNHCR 2007c, 257).
Women: (especially if victim of serious trauma), for example female-headed
households (unaccompanied females), widows, and young girls (at risk of forced
marriage, or use as bounty for settling community disputes and trafficking). This can
also include women with foreign spouses or trying to live a western life-style.
Children: (at risk of forced labour and trafficking), especially if unaccompanied and
victim of serious trauma
Those born in exile who lack social networks at home
The elderly and the sick, especially if unaccompanied, as well as physically and
intellectually disabled persons.
Families of communists, who may be persecuted upon return37
Afghans in areas where they constitute an ethnic minority (for example Pashtuns in
the North)
37
This depends on the positions held in the Communist government, as some ex-Communists
have returned to take up government posts in the current regime, while others have been elected
to serve in the lower house of the Parliament.
Page 26 of 44
Individuals (for example journalists, human rights activists) voicing critiques of power
figures or thought to be associated with internationals (in Taliban-dominated areas)
Landowners (due to land-grabbing and land disputes)
The Politicisation of Durable Solutions for Afghan Refugees

The previous sections in this chapter have highlighted that Afghanistan, the Afghan wars
and resulting refugees have all been subject to politicisation by multiple players. Such
politicisation is not uncommon, but perhaps more clearly than in other refugee
situations, the humanitarian and political dimensions of the Afghan refugee situation
were mutually reinforcing (Zolberg et al. 1989, 154) with refugees, and humanitarian aid,
becoming pawns in the larger geopolitical struggle for regional and international
domination (Loescher 1993, 89 see also Schch 2008).
In such an environment, the likelihood of also politicizing durable solutions is very
high, with political considerations of host and donor countries overriding the welfare of
refugees. Thus, repatriation was always a preferred durable solution for the Afghan
refugee population. Muslim countries were able to provide assistance in support of their
religious brothers and Pakistan could exploit the situation to bolster its influence in the
region. Pursuing other kinds of durable solution would have simply sent a wrong
message at this time. Furthermore, western countries did not have to open their borders
in order to resettle a very large refugee population.
Local Integration hindered by political developments
While refugees found ways to settle down in their host countries through intermarriage,
obtaining local ID cards, or simply blending in with the local population in areas with a
similar ethnic background, neither Iran nor Pakistan encouraged local integration
officially (Turton and Marsden 2002), especially in view of the large numbers involved.
(UNHCR 2006d, 144). Pursuing an official integration policy would have contradicted
political stances of both Iran and Pakistan linked to regional politics, the resentment
among the local population (mainly out of competition over jobs and housing) and
refugees being easy scapegoats for many social ills, particularly growing terrorism in
Pakistan. Nevertheless, it is highly likely that a significant number of Afghans will
seek to remain in Pakistan and Iran, both as family groups and as single wage earners,
and that they will find increasingly inventive methods to circumvent any attempts at
stricter border control and policy scrutiny inside the country (Turton and Marsden 2002,
52).
Adelkhah and Olszewska (2007, 148) argue that by turning its back on its
traditional welcoming policy towards Afghans, Iran [but also Pakistan] is undoubtedly
closing itself into a trap of impossible choices. Afghans in Iran and Pakistan are part of a
long migration history predating forced migration, with strong communal ties in border
areas. Complete expulsion would attempt to reverse a historical trend and leave both
economies without the benefits of cheap Afghan labour and neglect the symbiosis
between cities and economies in the border region of all three countries. This is
compounded by a border that is, if not impossible, still difficult to control and where
Page 27 of 44
Afghans have managed to blend into the local population given that a significant portion
of the population of border regions [is] ... of uncertain identity (Adelkhah and
Olszewska 2007, 154).
In sum, by driving a stringent return policy instead of looking for alternative
solutions, both countries situate themselves in the same kind of schizophrenia as
western European societies... as its [Irans] economy depends on travellers and
immigrant labour, whether or not they are legal (Adelkhah and Olszewska 2007, 157),
something which is very true as well for Pakistan. Still, all this inconsistency (at times
tolerating Afghan refugees/migrants and other times harassing and expelling them) may
actually all be part of a strategy that allows both countries to benefit from the cheap
labour of Afghans while not committing to local integration or permanent settlement
(Monsutti 2004).
Resettlement
Even though resettlement did occur, and still does, there never was a concerted and
significant resettlement program (in terms of numbers), as it would have been impossible
to resettle such vast numbers of refugees altogether. The majority of all Afghan refugees
were resettled during the beginnings of the refugee crisis, when the refugees still fitted
the Cold War mold. Figures are hard to obtain, as resettlement was mainly based on
individual applications for asylum and later family reunification schemes; but the total
was likely under half a million (well under 10 percent of the total Afghan refugee
population).
Access to resettlement is still difficult; as with the fall of the Taliban many Western
countries started incentive programmes in order to induce the voluntary repatriation
of Afghan refugees, especially for those with pending asylum claims (Amnesty
International 2003, 11), with several halting the judgement of new incoming asylum
claims. In the 2006 UNHCR Projected Global Resettlement Needs Report, UNHCR did
not project resettlement needs for the majority of refugees in Pakistan and Iran, due to
little willingness among potential countries of resettlement to take in more refugees,
clinging to the belief that the situation in Afghanistan had changed for the better. UNHCR
found only Afghan refugees residing in Central Asia to have urgent resettlement needs,
due to their having served in the past Communist regime, or having fled more recently
from sectarian violence and persecution. Similarly, the remaining Afghan population in
India, being mainly of Hindu or Sikh origin, and some Ismailis, may also be unable to
return to Afghanistan.
This is consistent with UNHCR Pakistans stating that resettlement is available
only for a limited number of Afghan refugees [a tiny minority] for whom neither staying in
Pakistan nor returning to their homeland is a safe option, such as women at risk or
security cases.38 They go on to argue that the number of refugees UNHCR resettles
from Pakistan remains among the highest of any country, despite the obstacles created
by much more stringent security checks since the 11 September 2001 attacks on the
38
http://www.unhcr.org.pk/about.html) [accessed 4 January 2008]
Page 28 of 44
United States.39 This leaves family reunification or marriage with refugees who hold
legal documents in Western countries as a last option to resettle. Smuggling is also still
attempted, but cannot be considered a durable solution, at least not officially.
The fact that most Afghan refugees have integrated well into their countries of
resettlement could be a consideration favouring resettlement in the future. However, the
profile of refugees resettled in the late 1980s and early 1990s differs from the ones that
currently remain. The bulk of resettlement prior to 1996 was provided to Afghans from an
elite background, well educated and aspiring to a western life style. Many of the
refugees still in Iran and Pakistan do not fit this description.
According to UNHCR statistics, 1,829 refugees of Afghan origin were resettled
from their first country of asylum in India, the Islamic Republic of Iran, Kazakhstan,
Pakistan, the Russian Federation, the Syrian Arab Republic and Tajikistan in 2006
(UNHCR 2006b). Prior to this, for the period 1996-2005, UNHCR assisted in the
resettlement of 26,846 Afghan refugees from Azerbaijan, India, the Islamic Republic of
Iran, Kyrgyzstan, Pakistan, the Russian Federation, Tajikistan, Turkmenistan, and
Uzbekistan (UNHCR 2005). This results in a global figure of less than 30,000 Afghan
refugees being resettled over the past 10 years.
Given the deteriorating security situation in Afghanistan, the unwillingness of major
host countries to integrate refugees locally, and increasing political instability in the
biggest host country, Pakistan, circumstances may force the international community to
reconsider its stance on resettlement. Especially in Central Asian countries such as
Tajikistan and Turkmenistan, the relatively small remaining number of Afghan refugees
seem to consider resettlement as their preferred durable solution (see IRIN 2003, 2004).
Some members of the Afghan refugee population in India, albeit in small numbers, are
also in dire straits as they are currently not allowed to work and thus are reliant on
subsistence allowances from UNHCR, illegal work, or remittances.
Repatriation Preferred durable solution?
The Governments of Pakistan and Iran, the two countries most affected by the presence
of Afghan refugees, have long insisted on repatriation as the preferred solution (UNHCR
2006d, 144). One small and two major waves of repatriation can be identified over the
past years (see Figure 4), with ad hoc and trickle movements occurring throughout. The
first peak in repatriation occurred when the mujahideen fighters finally captured Kabul in
1992 (even though the departure of the Soviet Army in 1989 already initiated tentative
return into an on-going conflict). After smaller movements totalling about 200,000 (19891991), the year 1992 saw a huge surge of collective optimism which resulted in no fewer
than 1.2 million Afghans returning from Pakistan in six months assisted by an
extremely stretched UNHCR (Colville 1997, 6); nearly another million moved home in
1993 (see Figure 4). Afterwards repatriation began to trickle off as the Afghan population
watched different political groups struggle for control of Kabul.
39
http://www.unhcr.org.pk/about.html) [accessed 4 January 2008]
Page 29 of 44
Figure 4. Return to Afghanistan, 1989-2007

2,500,000
2,000,000
Total
1,500,000
Pakistan
Iran
1,000,000
Other
500,000
7
20
0
5
20
0
3
20
0
1
20
0
9
19
9
19
9
5
19
9
3
19
9
1
19
9
19
8
Source: UNHCR (Statistics of Population of Concern, various years) and UNHCR

(2007b)
A second wave of return to Afghanistan, albeit much smaller in numbers, occurred after
1996 when the Taliban captured Kabul, and some Afghans initially were overjoyed to be
rid of what they saw as warmongering mujahideen fighters. However, having already had
one bad experience with return and renewed replacement, Afghan refugees did not react
as swiftly as before. As the Taliban soon began to impose strict constraints on everyday
life (including the banning of movies, music and photography), return remained well
under one million between 1996 and 2001. However, the subsequent fall of the Taliban
government and the establishment of the AIA [Afghan Interim Administration] by the
Bonn Agreement of 5 December 2001 led, in turn, to the largest and most rapid return
movement of Afghan refugees ever, and the largest UNHCR assisted repatriation
programme in almost 30 years (Turton and Marsden 2002, 17). As noted earlier,
according to UNHCR (2007b), nearly 5 million refugees returned to Afghanistan after
2002 in this third wave, out of which nearly 3.8 were assisted (see Figure 5). Of these
refugees, the majority returned from Pakistan (over 3 million), 90% of whom were
assisted in their return, and in the first year of return 61% were recent refugees who had
been in exile for five year or less (Lumpp et al. 2004, 150). Exact figures on unassisted
return are difficult to gauge and hence less reliable.
Page 30 of 44
Figure 5. Assisted vs. Non-Assisted (voluntary and forced) Return to

Afghanistan (2002-2007)
4,000,000
3,500,000
3,000,000
2,500,000
Assisted
2,000,000
Non Assisted
1,500,000
Forced
1,000,000
500,000
0
Total
Pakistan
Iran
Central
Asia/Europe
Source: UNHCR 2007b

The return of Afghan refugees was facilitated through various measures, which
clearly had an impact on the massive return figures in the first years (see Lumpp et al.
2004 for a detailed discussion), although Turton and Marsden (2002, 33) conclude that
return was mainly linked to assistance packages offered (having both a symbolic and
material value), to great, but misplaced, expectations about the level and early impact of
international assistance, and to push factors from host states:
First, return was organised through a tripartite repatriation framework in the region
that was negotiated by UNHCR, the first agreement being signed between
Afghanistan, Iran and Pakistan in 2002. Non-neighbouring countries that also signed
such agreements were France, the UK and the Netherlands.
Second, UNHCR launched massive information campaigns and arranged Go and
See visits for elders and representatives of Afghan refugees and Come and Talk
visits for community representatives and authorities to refugee camps.
Third, the Afghan government issued a decree of dignified return that assured
exemption from prosecution, irrespective of political affiliation, for criminal offences
against the internal and external security of Afghanistanexcept crimes against
peace or humanity or war crimes (Lumpp et al. 2004, 157).
Fourth, the Bonn agreement allowed Afghan refugee populations to participate in the
state-building exercise by reserving seats for representatives of refugee communities
in the 2002 Emergency Loya Jirga that decided on the Transitional Administration,
and allowing refugees to vote in the 2004 Presidential Elections. Furthermore, the
National Solidarity Programme (NSP), aimed at rebuilding rural communities,
prioritised those districts with high return figures, and NSP community development
councils were asked to incorporate returnees into their ranks.
Page 31 of 44
In contrast to Pakistan, and more in line with other refugee repatriations, the
majority (66%) of refugees returning from Iran did so without assistance. This again
mirrors the long-standing discrepancy in assistance to Afghan refugees hosted by
Pakistan as opposed to those hosted by Iran. More interesting, however, might be that
nearly as many refugees were forcibly deported from Iran (700,000) as those that were
assisted to return (UNHCR 2007b). This makes the return from Iran nearly evenly
distributed between assisted, spontaneous and forced (see Figure 5).40
Official reports put the forced return (deportation for security reasons) from
Pakistan at a much lower figure (about 6,000 refugees). However, it is important to
recall, as noted earlier, that at least in the case of the closure of the South Waziristan
camps in Pakistan, UNHCR accepted the official story from Pakistan claiming that return
was voluntary in order to be able to assist the refugees, despite the fact that refugees
told a story of threats and military bulldozing down camps by force, and UNHCRs
suspecting that the return was forced.41 Thus, some of the spontaneous unassisted (but
also some of the assisted) return from Pakistan may very well have been of a forced
nature.
Finally, as noted earlier, several western countries have initated an active
encouragement of voluntary return. Several of them (for example Germany, Switzerland
and the United Kingdom) have set up programs in collaboration with the International
Organization for Migration (IOM) while others even deported Afghans from their territory
who enjoyed only so-called complementary protection42 (see McAdam 2007) rather
than full refugee status. According to UNHCR (2007b), between 2002 and 2007, over
2,000 Afghan refugees were deported from western countries such as the United
Kingdom (1,959), Germany (236), Norway (199), and the Netherlands (76; UNHCR
2007b, iii). Still returns from Central Asia and Europe did not figure much in the overall
return statistics, partially due to the small number of refugees residing there.
As UNHCR assisted the majority of the return from Pakistan, it has been rather
well documented (UNHR 2007b, Monsutti et al. 2006), with more scant information on
those returning from Iran:
A majority of returnees from both Pakistan and Iran had only lived for a short time in
exile (5 years or less)
A majority of refugees from Pakistan and Iran returned to urban destinations (about
40%), with one third (29%) of all refugees from Pakistan returning to Kabul province
40
In addition to refugees returning home, UNHCR also assisted nearly half a million internally
displaced persons (IDPs) to return to their homes (UNHCR 2007b).
41
Informal discussions with UNHCR in 2004 when a local organisation with which one of the
authors (Schmeidl) worked, helped UNHCR with receiving the returning refugees in Paktika
province.
42
In order to provide an alternative basis for eligibility for protection for individuals who fall outside
the specificity of the 1951 Refugee Convention framwork, a variety of complementary protection
mechanisms have evolved drawing on human rights treaties (for example the International
Covenant on Civil and Political Rights, the Convention against Torture, the Convention on the
Rights of the Child) or international protection standards (e.g., providing assistance to persons
fleeing from generalized violence). These mechanisms are not as binding as the Refugee
Convention.
Page 32 of 44
alone, and it is likely that most of the refugees from Iran went to the western city of
Herat
Ethnicity roughly reflects the overall divisions within Afghanistan, with Pashtuns
dominating (56%) followed by Tajiks (25%) and Hazara (8%). Even though many
Hazara (Shia) refugees sought refuge in Iran (about 43% of documented Afghans in
Iran), their returns are only 25.6 percent of the total UNHCR-assisted return figures
up to August 2005. The imbalance is due both to greater economic opportunities in
Iran for Hazaras, and perceptions of continued prejudice against Shias in
Afghanistan (Abbasi-Shavazi and Glazebrook 2006, 7).
Only slightly fewer women (5% difference) than men repatriate; this figure shows a
slight increase in the last year of repatriation.
The majority of all returnees are families, with only 19% being single individuals. For
Afghan refugees in Iran, this is different, with single men making up the majority of
the refugees that repatriated outside the UNHCR voluntary return process or being
deported (the latter indicating links between refugees and labour migration among
Afghans in Iran, see Monsutti 2006, 14).
It is important to consider the education and skill level of returning refugees from
Pakistan, as according to UNHCR, a majority (63%) were non-literate (UNHCR
2007b). Of the 19% with education, the majority (56%) only completed primary
education and another 20% secondary (18% are under the age of five, and hence
not at a school age). This is somewhat puzzling given that both UNHCR and Afghan
political parties actually had set up schools inside the Afghan Refugee Villages.
Similarly, 67% of returning refugees claim to have no skills, with the remaining 33%
being split being domestic workers (48%), agricultural workers (12%), private sector
workers (8%), carpet weavers (7%) and construction workers (5%). This highlights a
long-term neglect in providing adequate education and skills development to one of
the biggest and longest standing refugee population in the world.
Conclusion: Sustainability of Return and Other Solutions

Very few observers would disagree that the future of return to Afghanistan for the
remaining refugee populations in Iran and Pakistan (but also elsewhere) is questionable
at this moment. Monsutti (2008, 19) states the obvious, something, with which UNHCR is
also coming to terms, namely that full repatriation is neither feasible nor desirable.
Nevertheless, while the vast numbers of refugees who have returned tend to be used to
demonstrate success, less attention has been given to the entire question of
sustainability of return leading to internal displacement after return, the recycling of
refugees or returning refugees changing status to migrant workers (legal or illegal). How
otherwise, for example, is it possible that 500,000 more refugees returned to
Afghanistan in the first four years of repatriation than we thought to exist in the first
place (Kronenfeld 2008, 2), with 3.5 million still remaining? Figure 6 illustrates the
seemingly bottomless pit, the unlimited supply, of Afghan refugees.
Page 33 of 44
Figure 6. Afghan Return Movements to Afghanistan
4000000
3500000
3000000
2500000
Refugees
Returnees
2000000
1500000
1000000
500000
0
2000
2001
2002
2003
2004
2005
2006
Source: UNHCR 2007b

There are several issues that are worth considering in the search for durable
solutions for Afghanistans displaced population. First, the pull factors influencing the
return of Afghan refugees to Afghanistan currently are, and will remain, extremely
modest (UNHCR 2007a). In a 2004 report, UNHCR (2004a, 1) already found a clear
correlation between willingness to return and length of exile. Those who had left during
the Taliban years or shortly prior have returned home, those who left 10 years or more
prior, remain. Thus, a clear distinction needs to be made between those who have
repatriated until now and those who remain. For UNHCR, natural repatriation has come
to a halt, and future repatriation may depend entirely on push factors linked to asylum
country policies, treatment of refugees by host societies and states, as well as forced
removal.43 Encouraging Afghans who have lived for two and a half decades outside their
country - some of whom, in fact, may never have even set foot in Afghanistan to
repatriate may be a distinct challenge in the coming months and years (Margesson
2007, 4). A refugee who has lived long in exile often lacks necessary ties and networks
at home that tend to facilitate return. This explains why most (about 82%) of all refugees
in Pakistan may have no desire or intention to return home. Similarly, most Afghans in
Iran refuges and labour migrants do not intend to return to Afghanistan in the
medium term (Abbasi-Shavazi and Glazebrook 2006, 2).
Second, the absorptive capacity44 of Afghanistan, a country that so far has not
succeeded in rebuilding its state and rule of law, needs to be carefully considered, as it
is linked to sustainable return (see also Monsutti 2008, Amnesty International 2003).
While return can legitimise a new government (Turton and Marsden 2002), which
explains why Afghanistan was very open for refugee return, it can also place an undue
43
44
Email exchange with UNHCR official in Afghanistan, 27 January 2008

The authors acknowledge that absorption capacity is difficult to quantify.
Page 34 of 44
burden on the stability of a country that has not yet dealt with justice and reconciliation
issues. One needs to ask the very basic question of how well Afghanistan is able to care
for its current population, let alone returning refugees, when it is largely dependent on
external donor support (Maley 2007).
The problems for those refugees that have already returned home were discussed
earlier and include issues such as lack of livelihoods, unemployment and underemployment, lack of access to land or shelter, low living standards, but also rights issues
such as unresolved community conflicts or fear of persecution of minorities (for example
Hazara who are often more reluctant to return, or Pashtuns that were persecuted in the
North). Security is also increasingly becoming an obstacle, especially in the south where
fighting has prevented refugees from returning to their homes. The deteriorating security
situation in Afghanistan has already had an impact on repatriation trends, which have
declined notably since 2005 making the maintainance of sustainable repatriation
increasingly difficult (UNHCR 2007c, 256), and also more expensive (Margesson 2007).
Many refugees who are unable or unwilling to return to their original homes due to
lack of livelihoods, security and de-mining move to Kabul and other big cities, which are
already stretched beyond capacity in terms of infrastructure able to support residents. All
this has already caused renewed internal displacement as we explored earlier45 and a
recycling of refugees who were forced to return to Iran and Pakistan (mostly illegally) to
rejoin family members (in refugee camps) or re-enter as new migrant labourers.46 Some
returnees have complained bitterly that their expectations of integration assistance in
Afghanistan were disappointed, leading to their decision to return to Pakistan or Iran
(Turton and Marsden 2002). This return to renewed displacement needs to be seriously
addressed when considering durable solutions for remaining Afghan refugees. As
UNHCR sees return as depending on Afghanistans overall progress toward political
stability and security (UNHCR 2007a, 260), the outlook is not promising.
This leads to a third point, the definition of sustainability of return as well as the
question of for whom we are finding durable solutions refugees or host countries. As
noted earlier, return in a post-conflict context is frequently treated as a priority, often less
for the benefit of refugees than for the messages it sends about the change at home
(Black and Gent 2004, Turton and Marsden 2002). A recent briefing paper by the
Development Research Centre on Migration, Globalisation and Poverty (2005, see also
Black et al. 2004) asks the question - what makes return sustainable? While living
conditions (and accessing rights) may not be ideal in the country to which refugees
return, they may be sufficient if they are better than in the host countries that they left.
This can make the return of some Afghans sustainable, while not for others, especially
those who still remain at this point. The impact of refugees on the areas to which they
45
For example, several refugees returning from Iran (about 11,000 in the Western Region, mainly
Maslakh camp in Herat) are currently hosted in makeshift IDP camps around Herat or simple
squatter settlements in urban slums (for example Kabul).
46
While some Afghan refugees leave permanently due to a lack of livelihood, some tend to leave
temporarily only, such as during cold winter months as survival may not be possible in poor
neighbourhoods with lack of electricity and heating.
Page 35 of 44
return also needs to be part of the assessment. In Afghanistan, this is a mixed picture
with many refugees contributing positively to the reconstruction process, while, as
discussed earlier, the sheer number does put a strain on the Afghan state. Similarly,
staying put, may not be the best way to understand sustainability either, especially in
the Afghan context (Monsutti et al. 2006). The briefing paper concludes that two factors
crucially influence sustainability of return voluntariness and the return environment.
This should discourage host states from forcefully removing or deporting refugees, as
they are likely to come back (under whatever status). Greater support of returnees once
they have returned home should be emphasized, something for which UNHCR has tried
to gain support (Turton and Marsden 2002).
Fourth, the psychological and personal reasons why a majority of refugees in both
Pakistan and Iran (but also abroad) do not show a strong desire to return home need to
be better understood. As just discussed, we tend to focus on the situation in Afghanistan
rather than looking at the experience abroad (Monsutti et al. 2006), especially the
exposure to a semi-urban lifestyle, higher standards of living, better public services
(health, education) and employment opportunities, all making refugees more reluctant to
go home (UNHCR 2006d, Grare 2003, Turton and Marsden 2002). Long-term exile also
tends to impact on the notion of what home means to many Afghans. For those who
have lived a long time abroad (or were born there) may consider their host countries as
home, while also having a nostalgic feeling for their country of origin (watan)
Afghanistan (Turton and Marsden 2002). Thus, while the Afghan identify of refugees
may still be strong, they may prefer to live in Pakistan, Iran or abroad. Instead of asking
who is a refugee and who is not, a better question might be: what is best for Afghans
and their neighbours? as it is unrealistic, considering the extent of economic
interdependence, to expect that every last Afghan will permanently return (Kronenfeld
2008, 4).
Fifth, solutions for the remaining refugee population need to be seen within a wider
framework than protracted displacement only (see UNHCR 2004a, Monsutti 2008).
Mobility has long been part of life for Afghanistan and its people for a variety of reasons
(economic, political and cultural, UNHCR 2004a, Adelkhah and Olszewska 2008,
Monsutti 2008). The Afghan wars only added to the magnitude and urgency of the issue,
making migration (forced or voluntary) part of an Afghan survival strategy. Afghans have
learned to spread risk through strategic mobility (Monsutti 2008, Monsutti et al. 2006),
which is a reason why many families tend to have some members in Afghanistan, some
in Iran and/or Pakistan, some in the Gulf states, and yet others abroad in western
countries. One could argue that Afghans never put all their eggs into the durable
solutions basket and thus can benefit from economic remittances and existing networks
abroad. Hence, we should give up any idea that sustainability equates to immobility that it ideally involves anchoring people to their places of origin (Turton and Marsden
2002, 52). As noted earlier, although the identity of Afghans refugees is often still strong,
and some may even still have strong links to their original watan, strategic
considerations result in a preference to stay put or engage in a more translational
lifestyle such as moving between Afghanistan and Iran/Pakistan. Thus, it is best to treat
Page 36 of 44
refugees as rational decision-makers that calculate, within existing circumstances, the

cost and benefits of return, possibly more than initial exit (Turton and Marsden 2002,
Monsutti et al. 2006).
In light of the above, solving the Afghan puzzle of protracted displacement may not
lie with the traditional durable solutions approach within the UNHCR framework, but with
greater attention to economic and migration complexities. According to Monsutti (2008),
the traditional framework is too simple for the complex nature of the Afghan situation of
which UNHCR (2004a) has been aware. This led UNHCR (2004a, 4) to suggest that the
nature of cross border movements to and from Pakistan and Iran should be studied more
closely to identify patterns and modalities, specifically the multidrectionality of population
movements (Monsutti et al. 2006). Given the current scapegoating of refugees in Iran
and Pakistan, such analysis should also include an assessment of the contribution of
Afghan refugees and migrants to both countries as well as the challenges of mixed
population flows (refugees and migrants) to host states (Turton and Marsden 2002). The
push factors in the form of the refugee policies of the two major countries of first asylum
(Pakistan and Iran), but also countries of resettlement that currently dominate the
voluntary return to Afghanistan, especially need to be adressed. As Crisp (2004) notes,
local integration does not necessarily mean the awarding of citizenship in host countries.
Afghanistan, Iran and Pakistan should be encouraged and supported to work out the
terms of any such agreement on their own assisted by outside actors open to nontraditional solutions to protracted refugee situations (Kronenfeld 2008, 17). This may
include a transnational model that allows for more permanent and flexible labour
migration agreements between Afghanistan and its neighbours (but also countries of
resettlement), which can benefit the country of origin (through remittances) as much as
the host countries (through access to labour). If no alternative solutions are found,
Afghans will adapt and find their own, as they have done so in the past. This may
include continued (illegal) labour migration to both Iran and Pakistan and Afghan
refugees blending with local populations in Pakistan and Iran.
Last but not least, in addition to managing the flow of persons who are moving
back and forward for economic and social reasons; and responding to the wishes of
those Afghans with legitimate reasons for remaining in the asylum countries, and
identifying those with genuine needs for continuing international protection (UNHCR
2004b, 14), the return environment in Afghanistan should also not be neglected. This
includes overcoming poverty, dealing with the absence of security and rule of law, and
working on reintegration programs. If problems inside Afghanistan as well as in the
main host countries are not addressed, a worst-case scenario could be greater regional
destabilisation. It may be easier and cheaper, however, to prevent this, by seriously
addressing the issue of protracted Afghan displacement now.
References
Page 37 of 44
Abbasi-Shavazi, M. J. and Glazebrook, D. (2006), Continued Protection, Sustainable

Reintegration:Afghan Refugees and Migrants in Iran (Kabul: Afghanistan Research
and Evaluation Unit).
Abbas-Shavazi, M.J. et al. (2005), Return to Afghanistan? A Study of Afghans Living in
Mashad, Islamic Republic of Iran (Kabul: Afghanistan Research and Evaluation
Unit (AREU)).
Adelkhah F. And Olszewska, Z. (2007). The Iranian Afghans. Iranian Studies, 49:2,
137-165.
Afghanistan Independent Human Rights Commission AIHRC (2007), Economic and
Social Rights in Afghanistan II, 1 August 2007. http://www.unhcr.org/cgibin/texis/vtx/refworld/rwmain?docid=471f4a5b0, accessed 1 January 2008
Amnesty International (2003), Afghanistan: Out of sight, out of mind: The fate of the
Afghan returnees, 23 June 2003. ASA 11/014/2003. http://www.unhcr.org/cgibin/texis/vtx/refworld/rwmain?docid=3f12f8a92, accessed 22 February 2008
AREU (2007), Feature: Registration of Afghans in Pakistan. Afghanistan Research
Newsletter, 14, 2-3 (Kabul: Afghanistan Research and Evaluation Unit)
Azarbaijani-Moghadam et al. (2002), The Transition from Relief to Development from a
Human Security Perspective: Afghanistan. Unpublished report to the UN
Commission on Human Security, (New York: United Nations).
Baitenmann, H. (1990), NGOs and the Afghan War: The Politicisation of Humanitarian
Aid, Third World Quarterly, 12:1, 62-85.
Benard, Ch., and Khalilzad, Z. (1984), The Government of God: Irans Islamic Republic
(New York: Columbia University Press).
Black, R. and Gent, S. (2004), Defining, Measuring and influencing Sustainable Return:
The Case of the Balkans, (Brighton: Working Paper T7 Development Research
Centre on Migration, Globalisation and Poverty).
Black, R. et al. (2004), Understanding voluntary return (London: Home Office Online
Report 50/04)
Boesen, I. W. (1988), What Happens to Honour in Exile? Continuity and Change Among
Afghan Refugees, in Bo Huldt and Erland Jansson (eds), The Tragedy of
Afghanistan: The Social, Cultural and Political Impact of the Soviet Invasion
(London: Croom Helm), 219-239.
Centlivres, P (1993), A State of the Art Review of Research on Internally Displaced,
Refugees and Returnees from and in Afghanistan. (Report prepared for the
Planning Committee of the Fourth International Research and Advisory Panel
Conference on Forced migration (IRAP). Oxford, January 1994).
Centlivres, P. (1987), Gestion administrative et organisation politique des camps, in .
Gille (ed.), Les Rfugis afghans (Paris: Amiti Franco-Afghane), 18-23.
Centlivres, P. (1989) L'Iran, l'autre pays d'accueil, Afghanistan Info, 24, 12.
Centlivres, P. and Centlivres-Demont, M. (1988a), Hommes dinfluence et hommes de
partis: Lorganisation politique dans les villages de rfugis afghans au Pakistan,
in Erwin Grtzbach (ed.), Neue Beitrge zur Afghanistanforschung (Liestal:
Stiftung Bibliotheca Afghanica), 29-43.
Page 38 of 44
Centlivres, P. and Centlivres-Demont, M. (1988b), The Afghan Refugees in Pakistan: A

Nation in Exile, Current Sociology, 36:2, 71-92.
Centlivres, P. and Centlivres-Demont, M. (1988c), The Afghan Refugees in Pakistan: An
Ambiguous Idenity, Journal of Refugee Studies, 1:2, 141-152.
Centlivres, P. and Centlivres-Demont, M. (2000), State, National Awareness and Levels
of Identity in Afghanistan from Monarchy to Islamic State, Central Asian Survey
19:3/4, 419-428.
Centlivres-Demont, M. (1993), Afghan women in peace, war and exile. In Myron Weiner
and Alij Banuazizi (eds.) The State and Social Transformation in Afghanistan, Irak
and Pakistan (Syracuse, NY: University Press)
Cheema, Pervaiz Iqbal. 1988. Impact of the Afghan War on Pakistan, Pakistan Horizon
41:1, 23-45
Collective for Social Science Research (2006), Afghans in Pakistan: Broadening the
Focus (Kabul: Afghanistan Research and Evaluation Unit).
Colville, Rupert (1997), The Biggest Case Load in the World, Refugees, 108, 3-9.
Connor, Kerry M. (1987), Rationales for the Movement of Afghan Refugees to
Peshawar, in Grant M. Farr and John G. Merriam (eds), Afghan Resistance: The
Politics of Survival (Boulder: Westview Press), 151-190.
Crisp, J. (2004), The Local Integration and Local Settlement of Refugees: A Conceptual
and Historical Analysis, (Geneva: New Issues in Refugee Research Working
Paper No.102, UNHCR Evaluation and Policy Analysis Unit.)
Development Research Centre on Migration, Globalisation and Poverty (2005), Defining,
Measuring and Influencing Sustainable Return (Brighton: Briefing Paper No.3,
Development Research Centre on Migration, Globalisation and Poverty).
Dupree, N. H. (1987), The Demography of Afghan Refugees in Pakistan, in Hafeez
Malik (ed.), Soviet-American Relations with Pakistan, Iran and Afghanistan
(London: Macmillan), 366-394.
Dupree, N. H. (1988), Demographic Reporting on Afghan Refugees in Pakistan, Modern
Asian Studies, 22:4, 845-865.
Dupree, N.H. (1992), The present role of Afghan women and children (The Hague:
Studies and Evaluation Papers 7, Bernard van leer Foundation,)
Edwards, D. B. (1986), Marginality and Migration: Cultural Dimensions of the Afghan
Refugee Problem. International Migration Review 20:2, 313-325.
Eighmy, T. H. (1990), Afghanistans Population Inside and Out (Islamabad: Office of the
A.I.D. Representative for Afghanistan).
Farr, G. M. (1985), The Effect of the Afghan Refugees on Pakistan, in C. Baxter (ed.),
Zias Pakistan: Politics and Stability in a Frontline State (Boulder: Westview Press),
93-110.
Glatzer, B. (1998), Being Pashtun - Being Muslim: Concepts of Person and War in
Afghanistan, in Bernt Glatzer (Ed) Essays on South Asian Society: Culture and
Politics II. (Berlin: Arbeitshefte/Working Paper 9, Zentrum Moderner Orient), 83-94.
Page 39 of 44
Glatzer, B. (2001), War and Boundaries in Afghanistan: Significance and Relativity of

Local and Social Boundaries. International Journal for the Study of Modern Islam,
2001(3), 379-399.
Glatzer, B. (2002), The Pashtun Tribal System, in G. Pfeffer and D.K. Behera (eds.)
Concept of Tribal Society, (New Delhi: Concept Publisher), 265-282.
Glatzer, B. (1998), Is Afghanistan on the Brink of Ethnic and Tribal Disintegration, in
William Maley (ed.) Funamentalism Reborn? Afghanistan and the Taliban,
(London: Hurst& Co.; New York: New York University Press, 167-181).
Goodson, Larry P. (1998), Periodicity and Intensity in the Afghan War, Central Asian
Survey, 17:3, 471-488.
Government of Pakistan (2007), Registration of Afghans in Pakistan (Islamabad,
Pakistan: Ministry of States and Frontier Regions, National Database &
Registration Authority (NADRA) and United Nations High Commissioner for
Refugees)
Grare, F. (2003), The Geopolitics of Afghan Refugees in Pakistan. in Stephen J.
Stedman and Fred Tanner (Eds.) Refugee Manipulation: War, Politics, and the
Abuse of Human Suffering (Washington, DC: Brookings Institution Press), 57-95
Guillo A. et al. (1983), La guerre en Afghanistan. Modificateions des dplacemen ts
traditionnels et mergence de nouveaux types de circulations, in Micheline
Centlivres-Demont (Ed.) Migrations en Asie. (Berne: Socit Suisse d'ethnologie),
139-153
Gutman, R. (2008), How We Missed the Story: Osama Bin Laden, the Taliban, and the
Hijacking of Afghanistan (Washington DC: United States Institute of Peace Press).
Hanifi, M. J. (2000), Anthropology and the Representations of Recent Migrations from
Afghanistan, in E. M. Godziak and D. J. Shandy (Eds.), Rethinking Refuge and
Displacement: Selected Papers on Refugees and Immigrants, 8, (Arlington:
American Anthropological Association), 291-321
Hussain, R. (2005), Pakistan and the Emergence of Islamic Militancy in Afghanistan
(Aldershot: Ashgate).
Internal Displacement Monitoring Centre (IDMC) (2007), Afghanistan: Fighting in the
South Sets Off New Wave of Displacement, 22 December 2006.
http://www.unhcr.org/cgi-bin/texis/vtx/refworld/rwmain?docid=459e61d02,
accessed 1 January 2008
International Crisis Group (2003), Afghanistan: The Problem of Pashtun Alienation
(Brussels: ICG Asia Report No.62).
Jawad, N. (1992), Afghanistan: A Nation of Minorities (London: Minority Rights Group).
Khalidi, N. A. (1991), Afghanistan: Demographic Consequences of War, 1978-1987,
Central Asian Survey, 10:3, 101-126.
Koser, K. (2007), Internal Displacement in Afghanistan, Presentation at Seminar on
Internal Displacement in South Asia, Brookings Institution, (8 November 2007),
http://www.brookings.edu/speeches/2007/1108_afghanistan_koser.aspx,
Kronenfeld, D. A. (2008), Afghan Refugees in Pakistan: Not All Refugees, Not Always in
Pakistan, Not Necessarily Afghan? Journal of Refugee Studies 21, 1-21.
Page 40 of 44
Kunz, E.F. (1973), The Refugee in Flight: Kinetic Models and Forms of Displacement,
International Migration Review, 7:2, 125-146.
Lischer, S. K. (2005), Dangerous Sanctuaries: Refugee Camps, Civil War, and the
Dilemmas of Humanitarian Aid (Ithaca: Cornell University Press).
Loescher, G. (1993), Beyond Charity: International Cooperation and the Global Refugee
Crisis. (New York, New York: Oxford University Press)
Loescher, G. (2001), The UNHCR and World Politics: A Perilous Path, (Oxford: Oxford
University Press).
Lumpp, K. et al. (2004), Voluntary Repatriation to Afghanistan Key Features, Refugee
Survey Quarterly 23, 149-173.
Maley, W. (1998), Introduction: Interpreting the Taliban, in William Maley (Ed.),
Fundamentalism Reborn? Afghanistan and the Taliban (New York: New York
University Press), 1-28.
Maley, W. (2001), Security, People-Smuggling, and Australias New Afghan Refugees,
Australian Journal of International Affairs, 55:3, 351-370.
Maley, W. (2002), The Afghanistan Wars (New York: Palgrave Macmillan).
Maley, W. (2003), A New Tower of Babel? Reappraising the Architecture of Refugee
Protection, in Edward Newman and Joanne Van Selm (Eds), Refugees and
Forced Displacement: International Security, Human Vulnerability, and the State
(Tokyo: United Nations University Press), 306-329.
Maley, W. (2006), Rescuing Afghanistan (London; Hurst & Co.).
Maley, W. (2007), Talking to the Taliban, The World Today, 63:11, 4-6.
Mansfield, D. (2007), Economical with the truth: the limits of price and profitability in
both explaining opium poppy cultivation in Afghanistan and in designing effective
responses, in Adam Pain and Jacky Sutton (Eds.), Reconstructing Agriculture in
Afghanistan (Rugby: Practical Action Publishing), 2130-234.
Margesson, R. (2007). Afghan Refugees: Current Status and Future Prospects, 26
January 2007. (Washington, D.C.: Congressional Research Service (CRS) Report
for Congress). http://fpc.state.gov/documents/organization/80250.pdf, accessed 1
January 2008
McAdam, J. (2007), Complementary Protection in International Refugee Law (Oxford:
Oxford University Press).
Monsutti A. (2006), Transnational Networks: Recognising a Regional Reality (Kabul:
Afghanistan Research and Evaluation Unit).
Monsutti, A. (2005), War and Migration: Social Networks and Economic Strategies of the
Hazaras of Afghanistan (New York: Routledge).
Monsutti, A. (2008, forthcoming), Afghan Migratory Strategies and the three Solutions to
the Refugee Problem. Refugee Survey Quarterly, 27:1 (forthcoming)
Monsutti, A. et al. (2006), Afghan Translational Networks: Looking Beyond Repatriation
(Kabul: Afghanistan Research and Evaluation Unit)
Orywal, E. (1993-94) Afghanen in Deutschland: Eine Statistische Notiz zur
Migrationssituation, Afghanistan Info 34, 17-19.
Page 41 of 44
Orywal, E. (ed.) (1986), Die ethnischen Gruppen Afghanistans: Fallstudien zu

Gruppenidentitt und Intergruppenbeziehungen (Wiesbaden: Dr. Ludwig Reichert
Verlag).
Qassem, A. S. (2007), Afghanistan-Pakistan Relations: border controversies as counterterrorist impediments, Australian Journal of International Affairs, 61:1, 65-80.
Rahjo, M. A. (2007), Afghanistan: UNHCR Considerations for specific groups relevant to
the determination of refugee status. in Austrian Centre for Country of Origin and
Asylum Research and Documentation (ACCORD) (Ed.) Country Report
Afghanistan, 11th European Country of Origin Information Seminar (Vienna, 21-22
June
2007),
23-54.
http://www.unhcr.org/cgibin/texis/vtx/refworld/rwmain?docid=473451a31e, accessed 1 January 2008
Rashid, A. (2000), Taliban: Islam, Oil and the New Great Game in Central Asia (London:
I.B. Tauris)
Rubin, B. R. (1997), Arab Islamists in Afghanistan, in John L. Esposito (ed.), Political
Islam: Revolution, Radicalism, or Reform? (Boulder, CO: Lynne Rienner), 179-206.
Rubin, B. R. 2000. The Political Economy of War and Peace in Afghanistan. World
Development 28:10, 1789-1803.
Rubin, B. R. et al. (2001), Afghanistan: Reconstruction and Peacebuilding in a Regional
Framework (Bern: Swiss Peace Foundation, KOFF Peacebuilding Report 1/2001).
Ruiz, H.A. (2001), Afghan Refugees in Pakistan at Risk. Refugee Reports 22:7, 1-8.
Saikal, A. and Maley, W. (1991), Regime Change in Afghanistan: Foreign Intervention
and the Politics of Legitimacy (Boulder: Westview Press).
Saikal, F. H., and Maley, W. (1986), Afghan Refugee Relief in Pakistan: Political Context
and Practical Problems (Canberra: Working Paper, Department of Politics,
University College, University of New South Wales).
Schetter, C. (2003), Ethnizitt und ethnische Konflikte in Afghanistan (Berlin: Dietrick
Reimer Verlag).
Schmeidl, S. (1994), Ethnic Dynamics in Forced Migration: A Comparison of Afghan and
Palestinian Refugees, International Journal of Group Tensions 24:4, 333-360
Schmeidl, S. (2002), (Human) Security Dilemmas: Long-term Implications of the Afghan
Refugee Crisis, Third World Quarterly 23:1, 7-29
Schmeidl, S. 2007, The Emperors New Cloth: The Unravelling of Peacebuilding in
Afghanistan, Friedens-Warte Journal of International Peace and Organizations
1-2(2007), 69-86.
Schch, R. (2008, forthcoming), UNHCR and the Afghan Refugees in the early 1980s:
Between Humanitarian Action and Cold War Politics, Refugee Survey Quarterly,
27:1 (forthcoming)
Shahrani, N. (1995), Afghanistans Muhajirin (Muslim Refugee Warriors): Politics of
Mistrust or Mistrust of Politics, in E.Valentine Daniel and John Chr. Knudsen (eds),
Mistrusting Refugees (Berkeley: University of California Press), 187-206.
Sliwinski, M. (1989a), On the Routes of Hijrat, Central Asian Survey, 8:4, 63-93.
Sliwinski, M. (1989b), Afghanistan: The Decimation of a People, Orbis, 33:1, 39-56.
Page 42 of 44
Stapleton, B.J. (2007a), A means to what end? Why PRTs are peripheral to the bigger
political challenges in Afghanistan, Journal of Military and Strategic Studies, 10:1,
1-49. http://www.jmss.org/2007/2007fall/articles/stapleton.pdf, accessed 29
February 2008.
Stapleton, B. J. (2007b), Afghanistan at the Cross Roads in Austrian Centre for Country
of Origin and Asylum Research and Documentation (ACCORD) (Ed.) Country
Report Afghanistan, 11th European Country of Origin Information Seminar
(Vienna, 21-22 June 2007), 7-23 http://www.unhcr.org/cgibin/texis/vtx/refworld/rwmain?docid=473451a31e, accessed 1 January 2008
Strand, A. et al. (2004), Afghan Refugees in Iran: From Refugee Emergency to
Migration Management (Norway: CMI-PRIO Policy Brief, 16 June 2004.).
http://www.cmi.no/afghanistan/peacebuilding/docs/CMI-PRIOAfghanRefugeesInIran.pdf, accessed 1 January 2008
Tapper, N. (1983), Abd Al-Rahmans North-West Frontier: The Pashtun Colonisation of
Afghan Turkistan, in Richard Tapper (Ed.), The Conflict of Tribe and State in Iran
and Afghanistan (London: Croom Helm), 233-261.
Terry, F. (2002), Condemned to Repeat? The Paradox of Humanitarian Action (Ithaca:
Cornell University Press).
Turton, D. and Marsden, P. (2002), Taking Refugees for a Ride? The Politics of Refugee
Return in Afghanistan, (Kabul: Afghanistan Research and Evaluation Unit
(AREU)).
UNODC (2007). Afghanistan Opium Survey 2007, (United Nations Office on Drugs and
Crime, 16 November 2007).
UNHCR (2000), The State of the Worlds Refugees: Fifty Years of Humanitarian Action
(New York: Oxford University Press).
UNHCR (2004a), Challenges to Return, March 2004. http://www.unhcr.org/cgibin/texis/vtx/refworld/rwmain?docid=4231bc0d4, accessed 22 February 2008
UNHCR (2004b), Protracted refugee situations, Executive Committee of the High
Commissioners
Programme,
Standing
Committee,
30th
meeting
(EC/54/SC/CRP.14,
10
June
2004),
http://www.unhcr.org/excom/EXCOM/40c982172.pdf, accessed 4 January 2008
UNHCR (2005a), Statistical Yearbook: Trends in Displacement, Protection and
Solutions,
Statistical
Annex
II
http://www.unhcr.org/statistics/STATISTICS/464478a72.html, accessed 4 January
2008
UNHCR (2005b), Afghan Refugee Statistics. February 2005. http://www.unhcr.org/cgibin/texis/vtx/home/opendoc.pdf?tbl=SUBSITES&id=421316072,
accessed
15
February 2008
UNHCR (2006a), Refugees By Numbers (UNHCR / MRPI / B1 / ENG 1, September
2006). http://www.unhcr.org/cgi-bin/texis/vtx/refworld/rwmain?docid=462df1862,
UNHCR (2006b), UNHCR Statistical Yearbook 2006: Trends in Displacement, Protection
and Solutions, Statistical Annex,
http://www.unhcr.org/statistics/STATISTICS/478cda572.html, accessed 29
February 2008
Page 43 of 44
UNHCR (2006c), Rethinking Durable Solutions in The State of the World's Refugees
2006, (Oxford: Oxford University Press), 129-199
UNHCR (2006d), Humanitarian Consideration With Regard to Return to Afghanistan,
May 2006. http://www.unhcr.org/cgibin/texis/vtx/refworld/rwmain?docid=4561c4094, accessed 22 February 2008
UNHCR (2007a) Afghanistan, in UNHCR Global Appeal 2008-2009. Geneva: UNHCR.
http://www.unhcr.org/publ/PUBL/474ac8e00.pdf, accessed 4 January 2008, 260263
UNHCR (2007b), Statistical Overview of Afghan Refugee Population in Pakistan, Iran
and Other Countries, Returned Afghan Refugees from Pakistan, Iran and NonNeighboring Countries, IDP Population Movements, Reintegration Activities and
Extremely Vulnerable Individuals (EVIs) Program (2 Jan 31 Oct. 2007),
Operational Information, Monthly Summary Report October 07, (Kabul:
Operational Information Unit)
http://www.aims.org.af/services/sectoral/emergency_assistance/refugee/unhcr_su
mmaries/oct_07/summary1.pdf, accessed 3 January 2008
UNHCR (2007c) South-West Asia, in UNHCR Global Appeal 2008-2009 (Geneva:
UNHCR), 256-260. http://www.unhcr.org/publ/PUBL/474ac8df11.pdf, accessed 24
January 2008
Westad, O. A. (2005), The Global Cold War: Third World Interventions and the Making of
Our Times (Cambridge: Cambridge University Press)
Wilkinson, R. (1997), An Iranian Surprise, Refugees (Focus Afghanistan) 108, 14-15
Zolberg, et al. (1989), Escape from Violence: Conflict and the Refugee Crisis in the
Developing World (New York: Oxford University Press)
Page 44 of 44

Afghan Refugees

Uploaded by

Copyright:

Available Formats

Afghan Refugees

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Afghan Refugees

Uploaded by

Copyright:

Available Formats

Finding Durable Solutions in Contested Transitions

The Case of the Afghan Refugee Population

Susanne Schmeidl (Griffith University)

2008 ISA Presentation (Schmeidl, Maley)

Figure 1. Afghan Refugee Population, 1979-2006

Source: UNHCR (Statistics of Population of Concern, various years)

2008 ISA Presentation (Schmeidl, Maley)

sustain an active neo-Taliban movement responsible for insurgency in southern

Afghan non-governmental organisations (NGO) provided another form of refugee

Afghanistan in order to send a message to Washington of some of the risks of pushing

http://www.unhcr.org/excom/40111aab4.html accessed 29 February 2008

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

partnerships with development organisations to ensure the successful reintegration of

Source: UNHCR Global Appeal 2007a

Email exchange with UNHCR official in Afghanistan, 27 January 2008

2008 ISA Presentation (Schmeidl, Maley)

Interview with UNHCR official, Kabul, Afghanistan, 11 November 2007

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

Figure 2. Documented Afghans in Pakistan by Ethnicity (2007)

http://www.refugees.org/countryreports.aspx?id=2001 [accessed 24 January 2008]

2008 ISA Presentation (Schmeidl, Maley)

Source: AMAYESH-II registration exercise (Iran 2005/6)

2008 ISA Presentation (Schmeidl, Maley)

There are various reasons why documenting IDPs in Afghanistan is difficult:

2008 ISA Presentation (Schmeidl, Maley)

humanitarian challenge (IRIN, 20 November 2007).23 The Internal Displacement

Afghanistan: Conflict-affected displacement major humanitarian challenge - Afghan Red

2008 ISA Presentation (Schmeidl, Maley)

landmines and a dearth of livelihoods (see Afghanistan Independent Human Rights

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

The Politicisation of Durable Solutions for Afghan Refugees

http://www.unhcr.org.pk/about.html) [accessed 4 January 2008]

2008 ISA Presentation (Schmeidl, Maley)

http://www.unhcr.org.pk/about.html) [accessed 4 January 2008]

2008 ISA Presentation (Schmeidl, Maley)

Figure 4. Return to Afghanistan, 1989-2007

Source: UNHCR (Statistics of Population of Concern, various years) and UNHCR

2008 ISA Presentation (Schmeidl, Maley)

Figure 5. Assisted vs. Non-Assisted (voluntary and forced) Return to

Source: UNHCR 2007b

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

Conclusion: Sustainability of Return and Other Solutions

2008 ISA Presentation (Schmeidl, Maley)

Figure 6. Afghan Return Movements to Afghanistan

Source: UNHCR 2007b

Email exchange with UNHCR official in Afghanistan, 27 January 2008

2008 ISA Presentation (Schmeidl, Maley)

2008 ISA Presentation (Schmeidl, Maley)

refugees as rational decision-makers that calculate, within existing circumstances, the

2008 ISA Presentation (Schmeidl, Maley)

Abbasi-Shavazi, M. J. and Glazebrook, D. (2006), Continued Protection, Sustainable