IRANICA
Herausgegeben von Alberto Cantera
und Maria Macuch
Band 29
2021
Harrassowitz Verlag · Wiesbaden
Yezidism
Between Continuity
and Transformation
Edited by
Khanna Omarkhali &
Philip Kreyenbroek
2021
Harrassowitz Verlag · Wiesbaden
Book jacket illustration: Yazidi sanctuary near the Syrian border, Sinjar Mountains
(Kurdistan, Iraq); © akg-images / François Guénet.
Bibliografische Information der Deutschen Nationalbibliothek
Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen
Nationalbibliografie; detaillierte bibliografische Daten sind im Internet
über https://dnb.de abrufbar.
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche
Nationalbibliografie; detailed bibliographic data are available in the internet
at https://dnb.de.
For further information about our publishing program consult our
website https://www.harrassowitz-verlag.de
© Otto Harrassowitz GmbH & Co. KG, Wiesbaden 2021
This work, including all of its parts, is protected by copyright.
Any use beyond the limits of copyright law without the permission
of the publisher is forbidden and subject to penalty. This applies
particularly to reproductions, translations, microfilms and storage
and processing in electronic systems.
Printed on permanent/durable paper.
Typesetting and layout: Kian Kahrom
Printing and binding: Memminger Medien Centrum AG
Printed in Germany
ISSN 0944-1271
eISSN 2750-4751
ISBN 978-3-447-11800-2
eISBN 978-3-447-39269-3
Contents
Abbreviations
VII
Philip G Kreyenbroek & Khanna Omarkhali
Introduction
XI
Part 1
The IS Attack and its Impact on
Yezidi Communities in the Middle East
Matthew Travis Barber (Chicago)
A Survey of the Political History of the First Year of the Yezidi Genocide
3
Sebastian Maisel (Leipzig)
Testing the Limits of Freedom: Syria’s Yezidis in the New Millennium
39
Leyla Ferman (Hannover)
Legal Justice and Yezidi Survivors
63
Khanna Omarkhali (Berlin)
Transformations in the Yezidi Tradition and the Position of Women
After the ISIS Attacks: an Interview with Ilhan Kizilhan
95
Eszter Spät (Budapest)
Displacement, Loss and Transformation: Yezidi Ritual Life in Iraq
107
Costanza Coppini (Berlin)
Yezidi Cultural Heritage in the Sinjar Region: an Assessment
135
Part 2
Diasporas and their Impact on Yezidism
Christine Robins (née Allison) (Exeter)
Fragile Yezidism, Hidden Strength
151
Khanna Omarkhali (Berlin)
The Complexity of the Notion of Honour in Traditional
Patriarchal Societies: the Case of Yezidism
181
Maria Six-Hohenbalken (Vienna)
We Never Buried Several Generations in One Country…
Today’s Challenges for the Armenian Yezidi Community
203
VI
Contents
Thorsten Wettich (Bremen)
Transformation Processes in the Yezidi Community in Germany
223
Sara Collot (Bologna)
The Last Ferman
251
Allison Taylor Stuewe (Arizona)
Iraqi Yezidi Survivors and the Making of the ‘Good’
Refugee Discourse in United States Media
285
Selected Bibliography of Works on Yezidis in English From 2000 to 2020
About the Authors
311
319
A Survey of the Political History of the
First Year of the Yezidi Genocide
Matthew Travis Barber
The Yezidi Genocide is the most catastrophic event in the modern history of the
Yezidi people.1 The few works on the Genocide that have appeared thus far have
mainly focused on the violence perpetrated against the Yezidis and no work has
attempted the construction of a general historical narrative covering the essential political dimensions of the Genocide and its aftermath.2 This chapter begins
to address this need by providing a basic chronology and description of pivotal
episodes in the political history of the Yezidi Genocide. Limitations of space
confine the present examination to roughly the first year of the Genocide, with
1
2
The content of this chapter is derived from the following sources: 1) first-hand
observation of the events described and direct engagement with the parties involved;
2) interviews and interactions with hundreds of informants over the past six years both
while conducting humanitarian and advocacy work in response to the Genocide and
during instances of focused research; 3) media publications from journalists and regional analysts; 4) reports from think tanks, government bureaus, UN agencies, and
other organisations; 5) knowledge produced by the Yazda documentation project whose
first team I appointed in Iraq when serving as director of the organisation in 2015. Much
of the material is taken from notes made as events unfolded over the course of the relevant years; citations of media reports are therefore only occasional. Extended quotations are from interviews I conducted, unless otherwise cited. Consistent with Iraq’s
patronymic naming conventions, subsequent references to persons are here made using
the individual’s first name, as an instance of the second name alone would indicate the
individual’s father or grandfather rather than the person being mentioned. The actions
against the Yezidis that began on 3 August 2014 are here referred to as genocide following their recognition as such by the UN and a plurality of national governments around
the world; references to this specific historical event are capitalised as a proper name.
Commentary on the Yezidi Genocide has often used the term ‘survivor’ with particular
reference to women who escaped sexual enslavement; in this chapter, however, the term
refers generally to anyone who survived the Genocide and does not specifically denote
female survivors of enslavement. As this chapter deals with a bounded period, the past
tense is used throughout, even when referring to individuals who are still living and
political entities or conditions that still exist. I would like to thank the Pozen Family
Center for Human Rights at the University of Chicago for supporting my work during
the period when this chapter was prepared.
Susan Shand 2018 has contributed an important examination of political aspects of the
American response to the crisis; the scope of her book, however, is limited to the initial
two weeks of the Genocide.
4
Matthew Travis Barber
some reference to the period preceding it. The first year of the Yezidi Genocide
was characterised by a series of failures in the provision of security as the threat
of violence loomed, followed by a pattern of inadequate responses once the disaster had begun. The events of this first phase of the Genocide, with the feelings
of neglect and betrayal that accompanied them, would shape the orientation of
the Yezidi community toward the involved regional, national, and international
powers over the subsequent years of the Genocide. A single chapter cannot detail every relevant aspect of the Genocide, nor can each included item be given
the extent of attention deserved. Nevertheless, this work introduces a number
of key features of the Genocide and the chronology will be expanded in both
scope and depth in future works.
I. The Stage Is Set: A Brief Summary of Circumstances and
Events Leading up to the Third of August
The toppling of Saddam Husayn in 2003 created a security vacuum in which
jihadist groups rapidly proliferated. One such group called itself ‘The Islamic
State in Iraq’ (ISI). Suppressed by US forces during the Iraq War years, jihadists
found a new opportunity to actively expand inside of Syria during the Syrian
Civil War that began in 2011. In 2013, ISI re-emerged in Syria and announced
its existence as ‘The Islamic State in Iraq and al-Sham’ (ISIS). ISIS proved immensely successful at co-opting and absorbing countless other rebel groups
fighting in Syria that either already had Islamist tendencies, were susceptible
to Salafi propaganda, or were militarily defeated and forced to join ISIS. Little
more than a year later, ISIS had achieved sufficient strength to re-enter Iraq, this
time not as an underground movement but with an army.
ISIS launched its campaign on 5 June 2014 and conquered Mosul on the tenth.
Working from the inside, its forces had already gained control of Falluja in January. At breakneck speed, ISIS was now able to expand its control over Sunni
Arab areas in the governorates of al-Anbar, Nineveh, Kirkuk, Salahuddin, and
beyond. On 29 June, ISIS announced its self-designation as a caliphate with its
leader, Abu Bakr al-Baghdadi, as caliph. At this juncture the organisation also
shortened its title to simply ‘The Islamic State’ (IS).
On 18 July, IS expelled the entirety of Mosul’s Christian population, stripping them of all their property and possessions.3 That IS allowed Christians
to leave with their lives, instead of giving them a choice between conversion
to Islam or death, was rooted in a theological commitment to classical Islamic
jurisprudence governing warfare and dealings with non-Muslim communities;
the fact that Islamic law categorised Christians as people-of-the-book granted
3
On this episode and the experience of Christians under ISIS/IS rule generally, see Barber 2016.
A Survey of the Political History of the First Year of the Yezidi Genocide
5
them certain rights that, in the eyes of IS leaders, should not be extended to the
Yezidis, a group not fitted with this label.
The genocidal intent that IS harboured toward those it considered deviant
became evident prior to the Yezidi Genocide. Beyond deliberations over classifications of non-Muslims, numerous atrocities were committed against Shi’is,
whose branch of Islam IS considered a threat to Islamic purity. On 12 June, IS
conducted a massacre of over one thousand Shi’i men enlisted in the Iraqi army,
machine-gunning them to death in trenches near Camp Speicher in Tikrit. Beginning 12 June and continuing for nearly a month, IS began targeting those members of the Shabak ethnoreligious minority whom it deemed to be Shi’is, attacking
and plundering eleven villages, killing or kidnapping over 135 people, and expelling the rest.4 Iraq’s Turkmens have both Sunni and Shi’i adherents and, as with
other Shi’i communities, the latter were violently targeted by IS. IS conquered
Tal ʿAfar, a Turkmen city located between Mosul and Shingal, on 16 June. Possibly
more than 125,000 Shi’i Turkmens fled the city,5 many on foot, and streamed into
Shingal as the Yezidis watched. Over the following week, IS conducted massacres,
kidnappings, and plunder in Shi’i Turkmen villages in Nineveh and Kirkuk.6
During this period, the Yezidis were also targeted directly. Prior to the IS
conquest of Mosul, Kurdish media sources had reported on a massacre of fifteen Yezidis (including women and children) in Ras al-ʿAin, Syria, on 28 May.7
Once IS had captured Mosul on 10 June, it ordered on the same day the mass
execution of all non-Sunni prisoners in the prisons and jails of Nineveh Governorate. This included a massacre at Badush Prison in Mosul, in which 670 or
more were killed, among them all the Yezidi inmates in the prison. Sunnis who
tried to intercede on behalf of their Shi’i or non-Muslim cellmates were massacred along with the rest.8 Scores of Yezidis were kidnapped in June, including
twenty-eight working as Iraqi border police near Baʿaj9 and civilians working
in Mosul. Yezidi community leaders secured the release of many of these by
paying ransoms totalling around one and a half million dollars,10 but the others
simply disappeared. The danger facing the Yezidi people was clear.
II. Security in Shingal on the Eve of the Genocide
‘Shingal’ can refer to the region, the district, the city, and the mountain. Shingal is a narrow, east-to-west-running mountain about sixty kilometres long
4
5
6
7
8
9
10
Barber 2014, p. 470; Human Rights Watch 2014 b.
Human Rights Watch 2014 b.
Hauslohner 2014; Human Rights Watch 2014 a; 2014 b.
EzidiPress 2014.
United Nations 2015, pp. 7, 9; Human Rights Watch 2014 c.
Human Rights Watch 2014 b.
Barber 2016, p. 470.
6
Matthew Travis Barber
that rises sharply out of its flat, desert surroundings. Its northern and southern
foothills were once dotted with Yezidi villages, but in 1975, Saddam had these
destroyed. He relocated their inhabitants into mujammaʿāt (‘collective towns’),
newly built in the desert at distances of up to eight miles from the mountain.
Saddam’s purpose was to render the people more controllable, more surveillable,
and to inhibit Kurdish resistance groups from taking shelter in the mountain
where local people might aid them. He accomplished this by disconnecting the
Yezidis from the mountain fastness that had provided them with protection for
centuries. Unbeknownst to Saddam, the relocation of the Yezidis into the plains
would also make them less defensible against a jihadist army nearly four decades
later. The north and south sides of the mountain each have a row of mujammaʿāt
running roughly parallel to the mountain. In addition to these large towns,
small Yezidi villages are scattered in the plains (some located even beyond the
mujammaʿāt), in the farmland around the base of the mountain, and tucked into
the foothills themselves. Some of these include historical villages that Yezidis
have slowly rebuilt in the post-Saddam era. The urban centre of the district
is Shingal City, which sits at the base of the mountain’s south-eastern slopes.
Up until the Genocide, a small minority of towns had mixed Yezidi-Arab or
Yezidi-Kurdish Muslim populations. The zone of Yezidi habitation therefore
radiates out from the mountain; beyond its perimeter are scattered numerous
Arab villages. The entire system described here can be referred to as the Shingal
region, i.e. the relevant portion of the Shingal District that is situated in relation
to the mountain. Though Yezidis were the majority in the Shingal region, demographic changes over recent years had made them a minority in Shingal City,
which was also home to growing populations of Sunni and Shi’i Kurds, Sunni
and Shi’i Turkmens, Arabs, and some Assyrians.
Shingal is one of a number of ‘disputed territories’ that the KRG claims should
be incorporated into the Kurdistan Region of Iraq (KRI). This claim rests on
a Kurdish ethno-nationalist argument asserting that any area with Kurdishspeaking inhabitants is an inalienable part of the Kurdish national homeland.
Problematically for the powers that would monopolise Kurdish nationalist discourse and the definitions of Kurdishness, premodern Kurdish identities were
far from monolithic11 and the Yezidis of Shingal had remained largely immune
to the influence of nationalist thought, partly because widespread illiteracy had
persisted in Shingal into the second half of the twentieth century. Typically,
therefore, Shingali Yezidis were unreceptive to attempts to convince them that
they needed to add a new dimension to their identity and be subsumed under a
Kurdish umbrella.12
11
12
See van Bruinessen 1994.
The frequently-posed question ‘are Yezidis Kurds?’ should be met with the fairly obvious
response that a Yezidi is a Kurd if identifying as Kurdish and a Yezidi is not a Kurd if he
or she does not identify as such. Yezidis from the Dohuk area who live directly under
KRG administration, are raised in Kurdish-majority areas, and grow up studying the
A Survey of the Political History of the First Year of the Yezidi Genocide
7
Since the fall of Saddam, security in Shingal had been controlled by the Kurdistan Democratic Party (KDP). The KDP worked early on to unilaterally impose
administrative authority over the Shingal region. Except for some token centralised forces, virtually all Kurdish Peshmerga forces in the country were partisan militias affiliated with either the KDP or the Patriotic Union of Kurdistan
(PUK). Peshmerga action, therefore, was often directed by party leadership apparatuses. The relevant apparatus in Shingal was what I call the ‘party centre system’,
which consisted of physical centres corresponding to a three-tier hierarchy of laq,
lijne, and rêkxraw. The system was crucial for maintaining political control, regulating Shingal’s economy, coordinating activities of the secret police,13 ensuring
that local security would be KDP-affiliated rather than answering to the central
government, and propagating Kurdish nationalism.
A district had a single laq, which was the chief headquarters and command centre for all other party centres in the district. Large towns within a district each
had a lijne that reported to the laq, and smaller towns, villages, neighbourhoods
within large towns, or outlying areas in the orbit of large towns had rêkxraws that
reported to their respective lijnes. A single lijne might direct dozens of rêkxraws.
Despite the use of Kurdish terms and some organisational differences, the party
centre system mirrored the Ba’th system of political centres that existed until 2003.
In Shingal, the system became economically important in how it served as
the dispensary of patronage. Using the portion of the Iraqi budget allocated for
the Kurdistan Region, the KDP bought loyalty in disputed territories through a
massive system of pay-outs whereby figures of influence received financial compensation in exchange for political party membership. Religious leaders and the
heads of tribes were especially targeted with patronage since the acquisition of
their loyalty would often result in the securing of the same on the part of those
under their authority. This strategy also rendered community leaders incapable
of critiquing power whenever the interests of their constituents were harmed,
since vocalizing objections to government policy could result in the loss of payments. Beyond leadership figures, patronage was also used to recruit thousands
of ordinary people as party members. The lijnes became centres of power in
that recipients of patronage would visit them to pick up their ‘salaries’. Loyalty
was also secured in other ways. By targeting the two largest sectors of public
employment — healthcare and education — the KDP was able to gain control of
the hiring for most government jobs. For many, therefore, membership with the
KDP became a prerequisite to the procurement of employment. For a number
of positions, a person would be required to visit their nearest lijne and procure a
13
curriculum of the KRG’s department of education are more likely to state that “Kurdish
is my nationality and Yezidism is my religion”. The same is usually the case for Yezidis
from Syria or Turkey. Yezidis from Shingal and Armenia, by contrast, are more like to
exclusively identify as Yezidi and to reject other ethnic labels.
‘Secret police’ here refers to a range of entities operating under the labels of Asāyīsh or
Parastin in areas under KDP control, both inside the KRI and in the disputed territories.
8
Matthew Travis Barber
document certifying their membership with the party. This was also required of
those seeking employment in the Kurdistan Region where many Yezidis would
travel to find better work opportunities.
The party centre system also became the entity that had ultimate authority
over all security in Shingal. This process began soon after the fall of Saddam.
To circumvent Coalition prohibitions of Peshmerga occupation of areas outside
the KRI, the KDP exploited a security loophole whereby they asserted a need
for security to protect their party centres. In this way, lijnes came to double as
pseudo military bases. Armed personnel would be stationed in lijnes and would
sleep there barracks-style. Peshmerga authorities also brought militarised vehicles with mounted weapons (something no local political faction possessed) into
Shingal and station them at party centers.
The secret police were closely integrated with the system and answered to the
laq. KDP secret police operated both weaponised and non-weaponised units;
the former were sometimes uniformed and maintained a visible public presence
at checkpoints while the latter were used for intelligence gathering. Secret police
in Shingal had their own facilities but also worked within and utilised the lijnes.
Detainments of political opponents of the KDP (which often resembled kidnappings rather than official arrests) sometimes involved bringing the arrestee
to a lijne for questioning or pressuring. Secret police could operate parallel to
the lijnes but the latter often had a role in directing the activities of the former.
The KDP also used its influence within the Nineveh governorate and national
government to guarantee that Iraqi army superiors in the Shingal region would
be Kurdish, a strategy that brought even the official armed forces under the authority of the laq through the mechanism of party loyalism.
The system also controlled all administrative positions from the very first
expression of post-Saddam governance in Shingal. The two highest-ranking
government positions in Shingal, the qāʾ immaqām (head of a district) and the
mudīr al-nāḥiya (head of a subdistrict, such as that which comprised the north
side of Shingal Mountain), were never elected by the people but were appointed
by gubernatorial and district councils that were controlled by the KDP. These
Yezidi officials were therefore selected and appointed by the party and accordingly functioned within a role of submission to the authority of the laq, which
was always headed by a Kurd and never a Yezidi. Yezidis were recruited into the
secret police and used to spy on fellow Yezidis, but top-level authorities of real
political and military consequence were always Kurdish.
The network of control in Shingal and its related monopoly on violence were
therefore indistinguishable from the political party apparatus. The party centre
system was the organism that held actual power in Shingal and, based in Shingal City, Laq #17 was the seat of this power. In 2014, the head of Laq #17 was a
Kurd named Serbest Bapiri.
When the Iraqi army collapsed in Nineveh upon the IS conquest of Mosul on
10 June, the Peshmerga moved to take full control of Shingal, supplementing the
A Survey of the Political History of the First Year of the Yezidi Genocide
9
forces they already had there. Serbest Bapiri was ultimately responsible for all
security. The Peshmerga generals responsible for troops in Shingal before the
Genocide were all Kurdish — none were Yezidi or from Shingal.
In terms of capacity to provide protection, the KDP Peshmerga in Shingal
came into an abundance of weapons and munitions in June. When the army collapsed in al-Anbar and Nineveh, thousands of Yezidis who were enlisted in the
Iraqi army returned home to Shingal, carrying their weapons. After their arrival,
the KDP secret police and Peshmerga worked together to disarm these soldiers
and round up their weapons. This seriously decreased the ability of Yezidis to
defend themselves when IS forces arrived and negatively impacted the local resistance that Yezidis conducted in the southern mujammaʿāt on 3 August.14
This action was repeated with the eleventh brigade of the third division of the
Iraqi army, which had been stationed in Shingal. When the Iraqi army dissolved
in the district, Laq #17 directed the Peshmerga to strip the eleventh brigade
of its weapons and equipment. The same thing was subsequently done to the
tenth brigade of the Iraqi army’s third division, which had battled IS at Tal ʿAfar
under the command of a general who went by the kunya15 Abu Walid. After
IS gained control of the city on the sixteenth, Abu Walid withdrew to the Tal
ʿAfar airfield southwest of the city and fought for several more days before being
forced to completely retreat. When the tenth brigade finally had to retreat from
the Tal ʿAfar area, its troops were only able to return to Baghdad via the KRI.
Obliged to pass through Shingal, the entire brigade was systematically looted,
being forced to hand over all weapons, ammunition, equipment, military vehicles, and uniforms — at Laq #17 in Shingal City. The troops were then sent back
to Baghdad via Erbil wearing civilian clothing.
Serbest Bapiri therefore presided over substantial acquisitions of new weaponry for security and military forces in Shingal. Further, the stripping of Iraqi
troops was not the sole source of new weapons for the Peshmerga; when the
14
15
In the case of Bakhdida (Qaraqosh), and other towns in the Nineveh Plain, the KDP
authorities went even further, going door to door and confiscating weapons owned by Assyrian civilians. Similar to what happened in Shingal, the Christian population of Nineveh
was promised that they would be protected by the Peshmerga but were later abandoned,
likewise with the Peshmerga pulling out prior to the arrival of the jihadists. The Peshmerga had well-fortified positions in Bakhdida and other places and could have resisted
the IS attack, but they withdrew without engagement. This occurred after the weapons
round-ups in July that were performed with official notices posted and circulated in those
towns. It seems incredible that the situation when northern Iraq faced its greatest security
threat since 2003 would have been chosen as the moment to disarm the local Assyrian
people; it is understandable how such actions fed the theories that sought to posit reasons
why the KDP could have been complicit in a plan to alter the demographics of minority
areas. Bayan Sami Abdul Rahman, who serves as the face of KDP PR for the United States,
later lied about Assyrian disarmament. See discussion (p. 74) and an example disarmament
notice with English translation (appendix 16) in Hanna/Barber 2017.
A nickname derived from the name of one’s oldest child or a personal quality possessed
by the bearer.
10
Matthew Travis Barber
Iraqi military dissolved, Kurdish forces also seized control of Iraqi weapons depots and snatched up all of the weaponry and ammunition. This included Kesek,
east of Tal ʿAfar, which housed the ‘regional munitions centre’ responsible for
providing weapons and ammunition to the second and third Iraqi army divisions and to military academies in Zakho and Slemani.
The appropriations described here represented a massive influx of weaponry
into the hands of Peshmerga in Shingal just prior to the Yezidi Genocide. Tragically, it is likely that many of these resources were sold on the black market.
Former secret police personnel stationed in Shingal whom I have interviewed
have recounted that this was commonplace. As Arabs were the most significant
buyers, it is probable that many of these weapons came into the possession of
IS. Eyewitnesses have attested to this occurring after the collapse of the Iraqi
border police in the Shingal region, under the auspices of corrupt officials. The
following is an excerpt from an interview with a former member of the secret
police in Shingal who spoke about smuggling and illicit weapons trading performed by Qasim Simo, a Yezidi from Kocho whom the KDP had installed as
head of the secret police in Nāḥiya al-Shamāl (the subdistrict encompassing the
towns and villages north of Shingal Mountain):
Qasim Simo controlled all the smuggling in Nāḥiya al-Shamāl. Weapons would
flow from Iraq into Syria. Smugglers could make good money because of the war
in Syria. A Kalashnikov that might normally cost $300 could be sold in Syria for
$400 or $500. If Dohuk secret police relayed the name of a smuggler to Snune
secret police, the Snune secret police were supposed to arrest him and take him to
Dohuk. But Qasim Simo would extort money from the smugglers instead. Qasim
Simo captured [name omitted], a smuggler. He told him, “If you want to keep
working and have a good life, you’ll need to pay me one third of each daftar [term
used in northern Iraq to denote $10,000]. Your name will stay in my desk and
will not go to Dohuk. If you don’t pay me, you’re going to disappear for one year”.
Qasim Simo would have us capture people and put them in the sun with no water for
hours. Afterwards they would quickly give him whatever information he wanted.
We would do this to smugglers who didn’t pay Qasim Simo the amount he wanted.
Qasim Simo knew about all the smuggling [in the area] but didn’t serve his people
[i.e. because selling weapons to Islamists in Syria ran against Yezidi security].
This was in 2012 and these weapons were going directly to ISIS. [ISIS declared
itself in 2013 but he is referring to the Islamists who later became consolidated
under the ISIS umbrella.] I’m sure our brothers and sisters were killed with the
same weapons being smuggled by Qasim Simo.
In June 2014, the Iraqi border police collapsed. On that day, Qasim Simo’s men
gathered all of the border police’s vehicles, weapons, and ammunition. It was
then about 4:00 a.m. Qasim Simo then told all of his secret police to leave and
come back at 8:00 a.m. When they came back, all of what they had gathered was
gone; he had sold it all to smugglers.
What seems incredible about the squandering of these resources is that it occurred amid the rapid encroachment of IS. Iraq’s border with Syria forms the
A Survey of the Political History of the First Year of the Yezidi Genocide
11
western and northern boundaries of the Shingal region. In the first half of 2014,
IS was in possession of sizeable portions of Syria immediately across that border.
As Syrian government authority weakened in the northeast over the course of
2012–2013, many Arab areas of the Jazeera became home to rebel groups that
competed with the Kurds and, once it emerged, IS had been able to quickly
extend influence within these towns. Before the Yezidi Genocide began, IScontrolled territory included a large swath of Hasaka Governorate directly west
of Shingal and stretching northward all the way to Tal Hamis, a mostly Arab
town less than twenty-five miles from Qamishlo.16
Inside of Iraq, the expansion of IS forces around Shingal was clearly observable for nearly two months prior to the Genocide, providing ample time to anticipate a potential attack and to plan for civilian evacuation. In fact, a number
of small attacks occurred on Yezidi villages southeast — and even north — of
Shingal in July. Tel Banat was attacked at least three times with both mid-size
machine guns and mortars, with some homes inside the town also hit by projectiles. In the two weeks immediately prior to 3 August, IS began taking over
Yezidi farmland south of the southern mujammaʿāt and flew their flags there.
They also targeted Yezidi farming areas adjoining the northern mujammaʿāt,
including an attack at Gohbal. KDP personnel stationed at the Gohbal lijne responded in a military vehicle that had a mounted heavy machine gun, but ended
up exiting it and fleeing on foot. The vehicle was seized by the IS fighters. Farmers north of Borek were later shot at on 1 August. Instances of kidnapping near
Yezidi towns also occurred before 3 August.17 The following interview excerpt
gives a sense of the climate during the week before 3 August:
My close friend from Ger Zerik, [name omitted], captured three Da’sh fighters
about one week before the Genocide at a Yezidi farm they had taken over about
one kilometre south of Ger Zerik. These were local Iraqis from the area who had
joined Da’sh. Da’sh fighters in those areas had been preparing for fighting by
making blinds and fortifications, and they were flying their flag, and everyone
could see them. So [name omitted] could have killed them, but he didn’t want to
create an outbreak in fighting, so he marched the three Da’shis to Ger Zerik. He
handed them over to KDP Peshmerga at a base inside of Ger Zerik. But the Peshmerga released them the same day. Everybody in Ger Zerik knew that the Peshmerga had released them. It actually made some people feel safer, because they
thought that it showed that Da’sh and the Peshmerga would not fight each other.
16
17
“The Islamic State has effectively used Kurdish-Arab tensions to rally Arab fighters in
the area of Tal Hamis to rise up against the YPG. This led to a significant portion of the
local Arab population declaring allegiance to the Islamic State”. Heras 2015, p. 9. As Tal
Hamis was not liberated by the YPG until 27 February 2015, the flight of Yezidis, various Kurdish operations, and early humanitarian activities in Shingal over the following
seven months all occurred adjacent to a nearby IS presence.
The first chapter of Nadia Murad’s autobiography (2017) recounts the kidnapping of
two of Kocho’s farmers in the early summer.
12
Matthew Travis Barber
Though the people described in this account were searching for signs of reassurance, most people did not feel safe. IS now held Tal ʿAfar to the east, Baʿaj to
the south, sections of Hasaka directly west, and Tal Hamis to the north: From
late June onward Shingal had been nearly surrounded by areas under IS control. Further, many had been keenly aware in June and July of the massacres of
Shabak and Shi’i Turkmens and Arabs, in addition to the kidnappings and the
slaughter of incarcerated Yezidis.
Noting these factors, many Yezidis observed that the prudent thing to do
would be to temporarily relocate to the KRI. For at least a month prior to the
Genocide, however, the KDP implemented a policy preventing this. While Yezidi
officials or individuals working in the KRI were still allowed passage between
Shingal and the KRI, families attempting to travel to the KRI were refused access
at the Suhayla checkpoint that regulated all movement to and from Dohuk.
On 2 August, this policy intensified amid the growing anxiety and families
were deliberately prevented from evacuating. As in its previous campaigns, IS
worked with local Arab agents to prepare for and plan the assault, a process that
inevitably involved information leaks. As rumours of the impending operation
circulated within Arab communities, some Arabs warned Yezidi friends and
colleagues that they expected that something was about to happen. KDP intelligence was aware of the situation and the heads of towns and villages made calls
to KDP security heads to ask whether the people should evacuate. Peshmerga
commanders, heads of security, and the Laq gave assurances to the people that
they would be protected and told them to stay in their villages. In cases where
families decided to evacuate anyway, secret police stationed at the entrances of
towns halted vehicles loaded with children and possessions and forced them
to go home. Some survivors have recounted that when being prevented from
evacuating, secret police or Peshmerga told them, “You should be ashamed for
trying to run away while we are here protecting you”.
III. The Islamic State Attack and the Peshmerga Withdrawal
IS mobilised its fighters at night on 2 August 2014. Two armies — one from Mosul and the second from Syria — were sent to simultaneously attack Shingal in
coordination. Key IS leaders who are known to have directly commanded forces
in the attack or who, because of their roles within the organisation, are likely
to have helped plan it included Amir Muhammed ʿAbd al-Rahman al-Mawla alSalbi, a Turkmen and religious scholar who helped found IS, was at least partly
responsible for anti-Yezidi fatwas, and was later selected as caliph of the Islamic
State after the death of Abu Bakr al-Baghdadi;18 Abd al-Rahman Mustafa al-Qa18
He has used the alternate name Abu Ibrahim al-Hashimi al-Qurayshi and also goes by
the nickname Hajji (or al-Hajj) ʿAbd Allah. Chulov/Rasool 2020; United Nations 2020.
A Survey of the Political History of the First Year of the Yezidi Genocide
13
duli, more commonly known as Abu ʿAlaʾ al-ʿAfri, who had lived part of his life
in Tal ʿAfar and was possibly Turkmen, had led jihadist activities in Iraq both
before and after 2003, had later become one of the most powerful men in the IS
organisation as ruler of its Syrian territory, and was also partly responsible for the
organisation’s religious positions against Yezidis;19 and Fadil Ahmad ʿAbd Allah
al-Hiyali, commonly known as Abu Muslim al-Turkmani, a Turkmen who was
as equally powerful as al-Qaduli in the IS organisation, administered IS territory
in Iraq, and headed the IS military council.20 These figures were Iraqi and some
had been imprisoned by the US during the Iraq War.21 Many other IS figures and
leaders were involved, but public information on their identities is scant; some
may be brought to light only through tedious future work comparing survivor
testimonies. Though foreign fighters participated in the attack, the bulk of the
jihadists were Iraqi and Syrian. These forces arrived after midnight.
In the early hours of the morning, when the jihadist advance was known
but before the attack began, the Peshmerga and KDP leaders began an unannounced withdrawal. Serbest Bapiri was one of the first to flee. KDP authorities
in most cases issued no calls to the civilians to evacuate, leaving people confused
as to what action to take.22 Chaos and pandemonium ensued. Many Yezidis on
the north side of the mountain only became aware of the situation after receiving calls from friends or relatives experiencing the attack on the south side.
The withdrawal involved virtually all KDP-affiliated security and military
personnel in the entire Shingal region — Peshmerga, secret police, and others. The withdrawal happened over a course of hours and different Peshmerga
forces therefore withdrew at different times. In almost no cases, however, did
they conduct any defensive action against IS; civilians were left completely unprotected. The Yezidi statement that became frequently heard after 3 August
— “The Peshmerga left without firing a single bullet” — was true for most all
locations around the mountain.
The withdrawal was ordered from Erbil23 and most of it was conducted in organised fashion with the Peshmerga transporting all weapons, equipment, and
military gear with them. Yezidis from nearly every town have given accounts of
19
20
21
22
23
An IS dissident’s narrative also confirmed al-Salbi’s role in overseeing the forced conversion of Yezidis in Shingal; the account was translated by Aymenn Jawad al-Tamimi 2019.
Al-Qaduli was also known as Abu ʿAli al-Anbari. He was killed March of 2016 by US
commandos in Syria. For the most comprehensive account of al-Qaduli, see al-Tamimi 2018.
Al-Arabia 2014; al-Tamimi 2016.
Some commentators on the rise of IS attempted to play up a supposed ‘former-Ba’thist’
aspect of the organisation; Craig Whiteside has published an important corrective to
this exaggeration that demonstrates that the organisation was decisively rooted in Salafi-jihadist ideology: Whiteside 2017.
Christine Van Den Toorn 2014 was the first to create a clear picture of the Peshmerga
withdrawal.
This has been affirmed in public interviews and private statements given by Peshmerga
commanders in Shingal; these will be detailed in a future work.
14
Matthew Travis Barber
beseeching the Peshmerga to leave behind weapons so that they might defend
their own families. In every case, this request was denied with the reason given
that it contravened official orders. This even happened at a Peshmerga training
camp in Domiz; when Yezidi recruits saw their superiors preparing to evacuate,
they pled with them to leave some weapons behind that they could use to defend
their communities, but they received the same refusal. As the Peshmerga were
leaving, Yezidis in the town of Zorava similarly begged for a few weapons to be
left for them; when this plea was rebuffed, several men tried to grab weapons
away from the departing forces. The Peshmerga fired on them, killing three
Yezidis and wounding at least two more.24
When the attack began, the southern mujammaʿāt bore the brunt of the assault. Local Yezidis organised themselves in defence of their own towns after
the Peshmerga fled, using whatever weapons they had. They continued to defend in several locations and made desperate calls to Peshmerga commanders,
begging for reinforcements to be sent to support them. None came. Not affiliated with the KDP command structure, a group of nine PUK Peshmerga in Ger
Zerik did not withdraw but remained and defended to the death. Their presence
in Ger Zerik remains a mystery as the PUK had no Sorani Kurdish forces stationed in Shingal at the time; it is possible that they were part of an intelligence
unit. A few individual members of the KDP Peshmerga did not leave as ordered
but remained to defend; these were mainly Yezidis who remained in their own
towns. Resistance was especially robust at the towns of Ger Zerik and Siba
Shaykh Khidr. Yezidi men at Siba Shaykh Khidr held out so well that they almost succeeded in weathering the IS attack and saving the town; the tide turned
against them only once they ran out of ammunition. That the Yezidis held out
as long as they did demonstrated how different the outcome would have been
had the Peshmerga stayed; even if some physical towns had not been saved, most
of the slaughter and enslavement that came to define the Genocide would have
been prevented if KDP forces had provided cover to civilians as they evacuated.
As Yezidi men shot back at attacking jihadists, residents of the mujammaʿāt began fleeing, as did those who lived in the scattered farming settlements between the
mujammaʿāt. Two smaller villages, Kocho and Hatimiyya, were surrounded and
captured with their inhabitants inside. As IS began to overcome resistance at the
mujammaʿāt, its forces began killing civilians in intersections, on the sides of streets,
and the on the main highway. IS forces also began rounding up women and girls
for mass abduction, having brought along empty trucks and buses for this purpose.
24
The three men killed were ʿAli Jabal, Yusuf Jabal, and Iyad Naif. One of the wounded
was Jamil Jabal. A young man from Zorava told me in an interview: “The Peshmerga
thought that for a soldier to give up his weapon is to give up his dignity. But they lost
their dignity the moment they left the people without protection. If I was a soldier, how
can I keep my dignity if I leave the people with no means of defence? If I am called a
soldier, a soldier is supposed to defend the people as much as possible”.
A Survey of the Political History of the First Year of the Yezidi Genocide
15
IS forces were like a sieve; they lacked sufficient manpower to either control
every town in Shingal or to capture over three hundred thousand people. IS
killed and captured thousands, but the greater part of the masses surged to the
southern slopes of the mountain and began ascending it — some via roads but
the majority by walking up little-used goat trails.
Yezidis inside Shingal City either fled up the mountain from the back of the
city or left via the city’s eastern entrance in an attempt to drive around the eastern end of the mountain to reach the north side where the road — in normal
circumstances — might take them to the KRI. When the jihadists reached the
city, they halted the stream that was pouring eastward out of the city, began
to force people out of their vehicles, and separated them according to religion.
Sunni Kurds were told to go back to their homes, while Yezidis were taken away.
Though the elimination of the Yezidis and the quest to sexually enslave
Yezidi women were broader objectives, the immediate priority upon arrival in
the city was to first kill the Shi’is. Shingal City was home to a small population
of Shi’i Kurds and the jihadists targeted their religious centre, the Husayniyya
Sitt Zaynab, blowing it up with explosives the same morning. IS had informants
alert them to the movement of the Shi’i Kurdish imam and his family, whom
they intercepted while fleeing. In mid-August, I encountered the imam’s elevenyear-old nephew in Dohuk; when the family was intercepted, the jihadists shot
and killed the boy’s father, two uncles, and aunt in front of him and the other
children of the family. Later the same day, the jihadists continued in this vein
and began demolishing Yezidi shrines in the same manner.
On the north side of the mountain, the jihadists first reached Khanasor and
Hardan, which were respectively the westernmost- and easternmost-positioned
towns. In contrast with the harsher fate of these two towns, which became the sites
of the largest mass graves on the north side, many — but not all — of the inhabitants
of the northern towns located further in were able to evacuate. Roadside killings of
those fleeing and rapid executions of stragglers occurred in and around almost all
towns. Some on the north side who did not wish to leave Shingal or who were not
able to evacuate fled up the mountain as those on the south side had done.
As had happened in some of the mujammaʿāt, local Yezidi resistance also
coalesced around the main access points to the mountain, supported by a small
number of PKK guerillas that had recently come to Shingal in the weeks leading
up to the Genocide. This proved sufficient to keep IS forces from ascending (and
gaining control of) the mountain, though jihadists did slaughter many people
as they fled up the foothills. After their initial control over the Shingal region
had been secured, jihadists went door to door in many Yezidi towns killing the
elderly and any others who had not fled their homes.
16
Matthew Travis Barber
IV. The Roles of Muslim Neighbours in the Genocide:
Arabs, Turkmens, Kurds
The Yezidi Genocide was a local phenomenon on several levels. IS was largely an
Iraqi organisation, in terms of both its leadership and much of its rank and file
in Iraq. Neighbouring Syrians were also an important component. Even closer
to home, Arab tribes whose villages neighboured those of the Yezidis played a
major role in both the attack and in the ensuing plunder of Yezidi towns.
In Shingal as well in as campaigns elsewhere in Nineveh, there were various levels of participation by which local actors joined or assisted IS. Prior to
its attacks, IS would utilise local Arab agents to produce intelligence. In many
cases, IS seemed to already know who the leadership figures were and where to
find them upon arriving. Local agents also helped them identify and differentiate between locals of different religious affiliations, verifying the statements of
those claiming to be Sunni and pointing out the homes of Yezidis, Shi’is, and
Christians.25 There were also factions who were not among the ranks of IS fighters but who would join IS when its campaigns entered their areas. Finally there
were cases of outright affiliation with IS where tribal leaders would pledge allegiance (bayʿa) to Abu Bakr al-Baghdadi; such a scenario subverted the clarity
of a straightforward actor-non-actor (i.e. militant-civilian) distinction as the
affiliation effectively deputised every male of the tribe as a member of IS.
Of particular relevance were the Shammar and Jhaysh tribes, whose villages
are mainly to the north of Shingal, and the Khatuni and Mitaywit tribes whose
populations were mainly in Baʿaj and other areas south of the mountain. Some Mitaywit Arabs lived inside Shingal City, also. Some Shammar individuals joined IS
but tribe leaders rejected it and the tribe therefore did not affiliate. Shammar Arabs
later worked with PKK affiliates and played an important role in driving IS out of
a number of locations southwest of Shingal. In contrast, the Jhaysh affiliated with
IS and played a larger role in the Genocide. A video appeared in the week after 3
August in which the Jhaysh tribe leader Abu Musʿab, from Khazuga village north
of Borek, was seen in Mosul giving bayʿa to al-Baghdadi in the presence of IS commanders. In addition to Arabs, many Sunni Turkmens from the Shingal region
participated in the Genocide. Smaller numbers of Kurdish Salafis joined the ranks
of the jihadists; some female survivors have reported that IS used Kurdish members to communicate with them during the early abduction process.26
25 Another tactic IS used to determine whether a family was Sunni or Shi’i would be to
take the children into a room without any adults and have them demonstrate how they
performed ṣalāt (Islamic prayer).
26 This linguistic capacity was useful to IS since those Yezidi women and girls who have
not had high school educations typically do not speak Arabic. Women and girls who
did not already have Arabic usually ended up acquiring it after being purchased in slave
markets and placed in Arab households.
A Survey of the Political History of the First Year of the Yezidi Genocide
17
Hardan, a large town but not a mujammaʿa, serves as a potent example of the
sense of betrayal experienced by many Yezidis. Hardan is a mixed Yezidi-Muslim
area. The road from Hardan and its twin Yezidi village of Kharana to the mountain ran through the small Arab village of Ger Shabak. Next to Hardan was Khidr
Amin, a Turkmen village. These communities lived together as neighbours for
generations with a good relationship. Residents of Hardan later recounted noticing drastic changes in the behaviour and customs of Arab neighbours soon after
the IS conquest of Mosul. These included the quitting of smoking, the adoption
of ‘Islamic dress’, positive remarks about the new ‘caliphate’, and comments made
to Yezidis that they were kuffār (infidels, unbelievers).
On the morning of 3 August, as Yezidis began to realise that they needed to
flee the area, only a few Yezidi families made it out of Hardan before their Arab
neighbours closed the roads, entrapping the majority of Hardan’s inhabitants
ahead of the arrival of the jihadists. Locals therefore facilitated the enslavement
of Hardan’s women and girls. Yezidis of Gormiz, a small hamlet in the northeastern foothill of the mountain whose men fought vigorously to prevent the
jihadists from ascending to their homes, watched through binoculars as IS conducted a mass execution of Hardan’s men. Their bodies now fill several mass
graves at the intersection connecting Hardan to the main road.
Together, Yezidis and Arabs in Shingal had long practiced a tradition called
kirîv that involved performing the circumcision of a boy while he was held on the
lap of a man from another family, ensuring that a few drops of blood fell onto the
man’s robes. Upon the completion of this ritual, the man would thereafter assume
a godfather-like role vis-à-vis the boy; more significantly, this also formed a deep
bond of mutual responsibility between the two families that verged on brotherhood. One of the most-lamented aspects of the Genocide was the unanticipated
occurrence of Arabs joining IS and turning on their kirîvs; Yezidis would remark
that they had assumed that kirîv bonds would protect them when a day of danger
came and expressed their shock and dismay that this had not been the case.27
Another level of participation in the Genocide was the looting of Yezidi
homes, farms, vehicles, livestock, and other property. Plunder was a chief pursuit of the jihadists themselves, but many local Arabs who had not joined in the
assault also participated in looting upon the displacement of the Yezidis.
Amid the depravity of this situation, there were also Muslims who saved the
lives of Yezidis, sometimes at great risk to themselves. I have collected accounts
of Yezidis who were safeguarded and hidden by Kurdish Muslims until they
could make an escape to the mountain at night. One man even described to me
how a masked jihadist, whose face he never saw but whose accent was local, in27 Some might interpret the kirîv ritual as evidence of a harmonious environment not defined by sectarianism; I believe, however, that Yezidis so actively sought to secure kirîv
ties with Arabs precisely because of a recognition of the salience of sectarian boundaries
and an awareness of the potential for tensions to emerge.
18
Matthew Travis Barber
terceded on his behalf with an IS amīr who was about to kill him and convinced
the man to let him go. One case is known where an Arab man lost his life, killed
by jihadists after retrieving his kirîv family and being intercepted while attempting to transport them to safety. Some Arabs warned Yezidis ahead of time that
violence was coming and that they should take precautions or leave. One survivor described this to me in an interview: “My brother had a friend who was an
Arab policeman who told him, ‘They will kill all of you — try to leave and save
your family.’ This was even before Da’sh took control of Baʿaj and there were still
Iraqi police in Baʿaj. After the attacks, this policeman called my brother and said,
‘I hope you and your family are safe.’” These acts do not mitigate the enormous
responsibility borne by the Muslim communities that share in the culpability for
this Genocide, but they were genuine acts of moral courage and humane compassion. Probably many more of them occurred than will ever be heard of.
V. The US Response
Most Yezidis who fled up the mountain on 3 August had no sources of food,
water, or protection from the sun. Some ate grass or leaves to survive. Others
killed domesticated animals and ate the meat raw. Still others went more than
a week without eating. Many family members had to choose who to carry and
who to leave behind among those who could not walk — children, aged parents,
those with disabilities, and the ill. Pregnant women gave birth on the rocky
slopes of the mountain; others had miscarriages. Feet were cracked and bloody,
skin was burned by the sun, and mothers used their own saliva to try and keep
little children alive. As scores of children and elderly Yezidis were dying in the
August heat, the US began performing airdrops of food and water on 7 August,
the fifth day of the Genocide. US airstrikes on jihadist forces in Shingal did not
occur until two days later, a full week into the Genocide.
The slow response was rooted in US policy objectives of the time, which were
twofold: first, to avoid a course of action that would appear to contradict President Obama’s objective of ending US military engagement in Iraq, and second,
to force the increasingly despotic Maliki regime into moving back toward certain democratic norms while compelling the Kurds and Baghdad to work with
each other in line with the ‘united Iraq’ policy.
Frustrated with Baghdad’s increased sectarianism, which had alienated the
country’s Sunnis (many of whom were now increasingly turning to IS as an alternative to the exclusionary Shi’i-dominated central government), the US had
deliberately allowed pressure to mount against Prime Minister Maliki. Before the
IS conquest of Mosul, as jihadists were building up forces and increasing their
control in Falluja, elsewhere in Anbar, and parts of Nineveh, Iraqi officials made
repeated requests to the US administration from March through May (including
A Survey of the Political History of the First Year of the Yezidi Genocide
19
during a 16 May conversation between Maliki and Vice President Joe Biden) for
strikes conducted by planes or drones. The US refused. The counsel offered by
one analyst seemed to reflect the rationale of the administration: “US military
support for Iraq could have a positive effect but only if it is conditioned on Maliki
changing his behavior within Iraq’s political system. He has to bring the Sunni
community back in, agree to limits on his executive authority and agree to reform
Iraqi security forces to make them more professional and competent”.28 As IS began making rapid gains across the country in June, many Iraqis blamed Maliki for
policies of discrimination and demands for him to resign proliferated swiftly. The
US administration avoided issuing a direct call for Maliki to step down, instead
making numerous statements on the need for a new, unified government that
could overcome divisions; by late June, however, US officials clearly perceived
Maliki’s prospective departure as a prerequisite for unity.29 After US lawmakers
urged Obama to make Maliki’s resignation a condition for greater US assistance
against IS, Maliki responded that he would not cooperate with such a demand.30
In an update on the situation in Iraq given on 19 June, the president stated:
It’s not our job to choose Iraq’s leaders… But I don’t think there’s any secret that
right now at least there is [sic] deep divisions between Sunni, Shia and Kurdish
leaders. And as long as those deep divisions continue or worsen, it’s going to be
very hard for an Iraqi central government to direct an Iraqi military to deal with
these threats. And so we’ve consulted with Prime Minister Maliki, and we’ve said
that to him privately. We’ve said it publicly that whether he is prime minister, or
any other leader aspires to lead the country, that it has to be an agenda in which
Sunni, Shia and Kurd all feel that they have the opportunity to advance their
interests through the political process.31
It was clear that the conversation had moved on from obligatory changes in
Maliki’s behaviour to the complete abdication of the prime minister, but even as
the nature of what was wanted from Maliki shifted, the policy of maintaining
pressure on him remained the same; as such, the IS conquests were allowed to
proceed largely unopposed. Nevertheless, the IS advance supplied insufficient
leverage, as Maliki valued the retention of his position in Baghdad over the fall
of Sunni and minority districts to IS. The US was also pressuring the Kurds
and Baghdad to cooperate with each other in the response to IS; though Maliki
finally ordered the Iraqi air force to support the Peshmerga on 4 August,32 he
continued to reject calls for his resignation.
The objective of forcing Maliki to give way to a government that would be
more inclusive of Sunnis therefore came at the cost of allowing a Sunni-led genocide of Yezidis to begin. The US approach involved the complete absence of
28
29
30
31
32
Gordon/Schmitt 2014.
Zengerle/Spetalnick 2014.
Homeland Security News Wire 2014.
Obama White House Archive 2014 a.
Associated Press 2014.
20
Matthew Travis Barber
any steps to brace Shingal for the impending onslaught. On 7 August, Obama
announced that he was authorizing airstrikes in Iraq for two reasons: to avert
the potential genocide of the Yezidis and to protect US personnel in Erbil. The
first strikes occurred on the eighth, targeting jihadists who were encroaching
on Erbil; strikes in Shingal then proceeded on the ninth. Even then, the US felt
that too robust an intervention would alleviate the pressure on Maliki. After the
strikes began, the president made it clear that engagement would remain limited,
stating that “there is no American military solution to the larger crisis in Iraq”
and continuing to call for the formation of “inclusive government”.33 On the
fourteenth, Maliki finally agreed to hand over the premiership to his successor,
after Iran had joined in calling for his resignation.
The US, therefore, began bombing IS not as a proactive measure, but as a last
resort. Despite years of Iraqi minorities begging for protection amid severe persecution — a problem that Obama had addressed at least twice in his capacity as senator34 — the US had taken no proactive measures to shield Shingal, even when the
Pentagon had detailed intelligence on IS movements in the area from June onward.
VI. YPG Intervention and the Corridor
Kurdish YPG35 forces in Syria took advantage of the US airstrikes that finally began on August 9 to mobilise southward from the Syrian border and fight through IS
lines until they reached the mountain. They then established an evacuation corridor
running from the mountain to the Syrian border. Yezidis began descending from
the mountain on the ninth, but the majority were evacuated from the tenth through
the twelfth. The number of Yezidis trapped on top of the mountain who passed
through this corridor was larger than most estimates that appeared in contemporaneous reports; the actual number could have exceeded one hundred thousand.
The significance of the corridor is difficult to overestimate. It saved the lives
of a large portion of Shingal’s Yezidi population. PKK guerrillas quite literally
carried elderly Yezidis down the mountain on their backs. The corridor also
established a link to Syria through which YPG forces could reinforce the ranks
of Yezidis fighting desperately in many remote corners of the mountain. This
changed the subsequent military outcome of the conflict. Though certain tensions did later arise between Yezidis and the PKK leadership in Shingal, this
initial action — coming directly after the KDP withdrawal — won many hearts
and minds within the community.
33
34
35
Obama White House Archive 2014 b.
Senator Barack Obama to Secretary of State Condoleezza Rice, 11 September 2007;
Senator Barack Obama to Secretary of State Condoleezza Rice, 26 September 2008.
The acronym denoting the People’s Protection Units, the main armed forces linked to
the PYD party, which is part of the larger network of PKK-affiliated groups.
A Survey of the Political History of the First Year of the Yezidi Genocide
21
VII. Hatimiyya’s Escape
Hatimiyya and Kocho are two smaller villages whose inhabitants belong to the
Yezidi Mantka tribe. In 2014, Husayn Berjis was the village head of Hatimiyya
and Ahmad Jaso was the village head of Kocho. On 3 August, both villages had
the similar experience of being captured with their inhabitants still inside, but
their situations ended with very different outcomes.
Kocho is the southernmost sizeable Yezidi village, located more than a thirteen-mile drive from the mountain. Its distance from the main Yezidi towns
made it especially vulnerable to capture early in the morning. When the Peshmerga stationed in Kocho left, some of the people managed to follow them; further evacuations were halted, however, when IS blocked the village’s exit routes.
Realising that attempting to flee could be as dangerous as any other option, all
that the inhabitants of Kocho and Hatimiyya could do was quietly wait as turmoil unfolded throughout the rest of Shingal.
On 6 August, an IS amīr visited the villages and gave the people a mandate
to convert to Islam, pledging to return on the tenth to hear their response. In
his discourse with the villagers, he made overtures of goodwill: “We’re not after
your money or your women; we just want to inform you that you’re now part of
the Islamic State and anyone in it must be Muslim”.
The residents of Hatimiyya and Kocho were not convinced. Though it was
difficult for the residents of small, isolated villages to form a clear picture of
what was transpiring, at least some word had reached them regarding the thousands of women who had already been abducted and the numerous massacres
that had occurred around the mountain.
Deciding that the safest course of action was to abscond before the IS amīr returned, the people of Hatimiyya waited for US airstrikes to provide them with an
opportunity to flee. The airstrikes never came. On the night of the ninth (the night
before the amīr’s next visit was expected), Husayn Berjis decided that the village
needed to attempt an escape — airstrikes or no airstrikes. Realizing that IS would
not allow the people to keep their lives if they chose to remain Yezidi, Husayn put
a plan in place, telling the villagers to run their small generators and turn on the
lights on their rooftops (where people slept in the summer) to give the appearance
that the people were comfortable and that everything was normal. He then directed everyone in the village to begin sneaking out under cover of the dark.
On the morning of the tenth, no one from Hatimiyya could be reached by
telephone and observers at first believed that Hatimiyya’s population had been
slaughtered. Within a few days, however, it was discovered that everyone in
Hatimiyya had successfully escaped to the mountain.
22
Matthew Travis Barber
VIII. The Kocho Massacre
Whereas Hatimiyya had performed a daring escape, the villagers of Kocho,
which was larger than Hatimiyya, felt helpless. Distance was a primary factor;
Kocho lay approximately twice as far from Shingal City than did Hatimiyya.
Moreover, after Hatimiyya’s night flight, IS increased security around Kocho,
rendering chances of escape even more remote.
IS did not act immediately following the expiration of their first ‘invitation to
Islam’, but delivered additional ultimatums regarding conversion. Many people
of Kocho did not expect to live through this situation, and they communicated
this to the outside world. On a daily basis during the occupation of the village,
I would call Yezidis inside of Kocho, confirm that they were still alive, publicly
report on the ultimatums to convert, and beg for a military response. People
in the village informed me that IS sentries would scamper for cover every time
a coalition airstrike occurred in the area, indicating that a Special Forces rescue operation could have effectively rescued Kocho’s inhabitants. US military
leaders were well aware of the situation in Kocho, as American Yezidis who
had travelled to Washington were meeting with US officials directly and urging them to act.36 Sadly, no action was taken. After a team of US Special Forces
visited Shingal Mountain and saw that the Yezidis entrapped on its summit had
been able to successfully evacuate via the YPG corridor, President Obama announced on 14 August that the siege of the mountain was ‘over’. In conjunction
with this, Pentagon officials cancelled plans for a potential rescue mission they
had been considering. Speaking from a plane where he was accompanying Defence Secretary Chuck Hagel, Pentagon Press Secretary and Rear Admiral John
F. Kirby was quoted as saying, “We get paid to think in worst-case scenarios,
that’s what we do. Getting on the mountain and seeing for ourselves that things
weren’t as bad as we thought was a pleasant surprise”.37 But the displacement
emergency on top of the mountain was not the only crisis in Shingal and the
president’s comments inadvertently conveyed a sense that the worst of the disaster had somehow passed. Yezidi leaders were furious and issued statements
asserting that the situation remained terrible and that tens of thousands of people were still trapped on the southern side of the mountain, many without food
or water, unable to ascend its steep slopes, and still besieged by IS; this message
was echoed by leading UN officials inside Iraq.38 In an article published the
same day that Obama made his optimistic comments, I reported the following:
“A number of people have told me that they are receiving calls from relatives
trapped inside besieged villages. They are calling for one purpose: to inform
36
37
38
The initial advocacy performed in Washington by US-based Yezidis immediately upon
the inception of the Genocide is profiled in Shand 2018.
Cooper/Shear/Nordland 2014.
Nordland 2014.
A Survey of the Political History of the First Year of the Yezidi Genocide
23
their families that they will soon be killed for refusing to convert to Islam”.39
The next day, my calls to Kocho went unanswered; those with whom I had been
in contact were dead.
On 15 August, IS personnel brought everyone in Kocho to the village’s
school, where they separated the women from the men, stripped everyone of
any valuables on their persons, and led the postpubescent boys and adult men
away. The males were transported in successive groups to various locations outside of the village and shot. Women and children could hear the gunfire from
where they were held before being transported away to Solagh, near Shingal
City, to be processed for the enslavement project. Another massacre followed in
Solagh, this time of middle-aged and older women whom the jihadists did not
wish to enslave. Among them was the mother of enslavement survivor and later
Nobel Peace Prize recipient Nadia Murad.
On the fifteenth, I reported that an estimated two hundred men had been
killed in Kocho. This figure was sadly conservative as it later became evident that
hundreds more had been killed.40 The announcement was met with disbelief and
incredulity, despite nearly two weeks of warnings and pleas for action beforehand.
The only men of Kocho to evade death were those already outside of the village on 3 August, those who happened to leave before IS blocked off the potential
routes of escape, and a handful who survived their bullet wounds and played dead
until they could sneak away. None of Qasim Simo’s family perished in the massacre; prior to the arrival of IS forces on the third, Qasim had quietly evacuated
his relatives from Kocho without notifying the other villagers that they should
leave. On 16 August, as rumours of the Kocho Massacre were spreading through
the community, a Yezidi community leader from the north side of the mountain
called Ahmad Jaso’s cell phone. A jihadist answered the phone and asked, “What
do you want?” The caller said, “I want to speak to my friend, Ahmad Jaso”. The
voice replied, “Ahmad Jaso is dead; I slaughtered him like a sheep”.
39
40
Barber 2014b.
No official count of those killed in Kocho has been established and reports have given
estimates of Kocho’s pre-massacre population ranging from 1,200 to 1,700 people. The
segment comprised of all postpubescent males and most of the middle-aged and older
women no doubt represented a large percentage of this population. For more on the Kocho Massacre, see the following reports, each of which has introduced new findings over
time: Rovera 2014; Amnesty International 2014; Kikoler 2015; Yazda 2016; Bor 2018;
Cetorelli/Ashraph 2019. From the last report in that list, note the following excerpt
(p. 9): “ISIS fighters, led by Abu Hamza, held multiple meetings with Ahmed Jasso and
other prominent men in Kocho. … Some of the village leaders told their families that the
men were trying to reach a resolution with ISIS, possibly by having the Yazidis give up
their possessions as the price for safe passage out of ISIS-controlled territory. Kocho is,
as far as is currently known, the only village in which such intensive negotiations occurred. Why remains unclear. Some have posited that the commander was from the area
and so was willing to enter into these negotiations. Some, emphasising the friendships
and economic partnerships that connected the various communities in Sinjar, suggested
that Arabs from neighbouring villages intervened in an attempt to save the Yazidis”.
24
Matthew Travis Barber
The Kocho Massacre did not deviate significantly from the general pattern
of IS violence throughout Shingal; the special attention sometimes afforded it
should not detract from the numerous other massacres that until now remain
neglected. The Genocide involved massacres in Yezidi towns, villages, and
open areas throughout the Shingal region and Kocho was not the only place
where men (and sometimes women) were lined up and machine-gunned next
to trenches. The Kocho Massacre, however, was notable for several reasons, the
first being its scale (it may have been the instance of the largest percentage of a
single community was wiped out), and the second being the protracted, observable period leading up to it when intervention could have taken place.
IX. The Coalescence of Early Resistance
In the wake of the Peshmerga withdrawal, a number of local resistance militias formed in Shingal. The most robust of these was the Shingal Resistance
Units (Yekîneyên Berxwedana Şengalê, abbreviated YBŞ). As soon as they had
brought their families to safety, many Yezidi men were impatient to join the fight
against IS. The KDP offered no avenue for this, but the PKK/YPG immediately
began training large numbers of Yezidi fighters at several bases in Syria. Yezidi
community leaders gathered eager volunteers, organised them into groups, and
began transporting them to Syria with the first wave leaving the KRI on the
sixth. The YBŞ became the most active and effective Yezidi resistance group,
often working alongside PKK guerrillas who maintained many fronts with IS
in northern, western, and southern areas of the mountain.
Jiwana tribal resistance formed around Sharfeddin, the largest Yezidi temple
in Shingal. Its surroundings are primarily Jiwana-inhabited and the area became an important locus of defence in the north-eastern part of the mountain.
Qasim Shesho and his nephew Haydar Shesho were the faces of this resistance;
the former was already a KDP member and the men loyal to him later became
organised into a KDP-affiliated militia while the latter, though a member of the
PUK, led an independent militia (the Shingal Protection Force, Hêza Parastina
Şingal, abbreviated HPŞ) until its near-complete disbandment in April 2015, reconstituted a year later in closer association with the KRG.
Many other militias coalesced around individuals who had rallied men together to defend different parts of the mountain on 3 August. Some of these
were later absorbed into the YBŞ or the KDP, while others have remained
unaffiliated up to the present. One example was Qasim Derbo, a poor Yezidi
mechanic with no military background who may have single-handedly saved
the mountain from being captured by IS. When the Peshmerga withdrew, one of
their vehicles broke down on the road that winds up the mountain from Shingal
City. The vehicle had a heavy weapon mounted on the back and the Peshmerga
A Survey of the Political History of the First Year of the Yezidi Genocide
25
abandoned it on the road. Later the same day, as the jihadists began pursuing
Yezidi civilians fleeing up the mountain, Derbo climbed into the vehicle and
aimed the gun at the IS forces, halting their ascent. The KDP later put Derbo on
their payroll and made him commander of some units who had gathered around
him. The KDP also erected a monument to the incident, building a replica vehicle and installing a giant portrait of Masoud Barzani next to it.41
When resistance formed in August, there was a sense of unified solidarity
in the struggle to defend against IS. It did not take long, however, for political
tensions to arise. Unaffiliated Yezidi militias that had sprung up around the
mountain had virtually no outside support in terms of both weapons and nonmilitary aid. US action to back the defenders lacked fervour; out of 170 US
airstrikes against IS in Iraq since 8 August, by 19 September only fourteen had
been conducted in Shingal.42 Rains began to arrive in October and food was
scarce especially for Yezidis whose defensive positions were located in remote
parts of the mountain. Following the Peshmerga’s recapture of Rabiʿa on 2 October, no reinforcements continued south to assist Shingal. Instead, KDP forces
focused later in the month on recapturing the oil-rich Zummar region. Outmanned by jihadists, Yezidi fighters lost control of Gohbal on 4 October. On
the same day, an ordinarily mute Mîr Tahsin Beg, the traditional head of the
Yezidi community, issued a statement containing rare criticism of the KRG and
central government for providing no ‘serious attempt’ to assist Yezidi resistance or pursue Shingal’s liberation.43 Yezidis claimed that weapons and supplies
shipped by Baghdad via Erbil were being withheld by the KDP; upon delivery
by helicopter to a base of KDP secret police on top of the mountain, Yezidi
militias were told that no aid would be shared with them if they refused to
identify as Peshmerga. Yezidis protested at the KDP base and hand-to-hand
clashes broke out on the eighteenth. IS launched a comprehensive renewed assault on 19 October, attacking Yezidi positions around the mountain. On the
twentieth, IS succeeded in cutting off YPG access to Syria, closing the corridor,
and retaking additional northern mujammaʿāt that Yezidis and PKK forces had
wrested from them. IS conducted repeated attacks on Sharfeddin that nearly
overwhelmed the Yezidi forces defending the temple. On 22 October, Shaykh
Khayri Murad Shaykh Khidr, leader of an unaffiliated militia from Siba Shaykh
Khidr, was killed in a southwestern area of the mountain. Reportedly inhibited
41
42
43
There was irony in that the monument memorialised a moment of heroism that itself
seemed to highlight the Peshmerga withdrawal. This fortunate incident happened because a Peshmerga vehicle suffered a breakdown and was abandoned; it would have otherwise been taken out of Shingal along with all other Peshmerga resources. The logic of
patronage allows such realities to be shamelessly ignored as long as the heroic figure is
co-opted. In this manner, an episode of negligence can be converted and presented as the
self-same group’s bravery and heroism.
BBC 2014.
Barber 2014a.
26
Matthew Travis Barber
by stormy weather, coalition airstrikes on attacking IS forces had remained seriously inadequate and the mountain was now completely besieged for the second
time. The mid-October IS offensive raged until airstrikes provided some relief
on the twenty-ninth and thirtieth, but the corridor remained closed. On 31 October, a Peshmerga force sent by Masoud Barzani arrived in Kobani, Syria, to
aid the IS-besieged Kurds there. The Peshmerga delegation delivered weapons
to the Kurdish fighters but did not assist them in combat. The plight of Kobani
stirred hearts throughout the Kurdish world, but to many Yezidis the delegation looked like a PR stunt. Yezidis were incensed to see Peshmerga travel some
350 miles to support a struggle in another country while Yezidi fighters were
locked in a precarious existential battle in IS-occupied Shingal.
On the same day that Peshmerga forces arrived in Kobani, a historic delegation of Yezidi religious and tribal leaders met in the White House with
the Obama administration, the culmination of a week-long series of meetings
with officials in various agencies across Washington. Among them was Husayn Berjis of Hatimiyya. The leaders had come from Iraq and Europe to plead
for real action against IS and real support for Yezidis, in particular, assistance
that would enable them to defend their homeland without having to submit to
assimilation by any of the external Kurdish political parties with aspirations
of ruling and annexing Shingal. The KDP was very concerned about what
the delegation might say to US officials, so it sent D.C.-based Kurdish party
members to meet with the participants privately and to feel out the delegation.
In this meeting, members of the delegation — who themselves belonged to the
KDP — stated that “we do blame you for what happened to us, but we will
not say this in our meetings with officials”. Nevertheless, the delegation broke
with party policy to submit to the State Department and White House a list of
written demands in the name of Mîr Tahsin Beg, one of which was assistance
for creating a legitimate and official local protection force, “under a unified
and joint leadership so that it will include all the segments of Yazidi society
in the Shingal region united under one umbrella, not apart from the state, but
within the Iraqi defence system”. Yezidis could not survive on their own and
were caught between would-be sponsors contending for their loyalty; they
hoped that the West would free them from this dysfunction by becoming their
direct sponsor and empowering nonpartisan security. In these meetings, US
officials mentioned a pre-existing idea of an ‘Iraqi national guard program’
within which the officials said they would support the inclusion of ‘a local
Yezidi component in Sinjar’. But the US never pursued an effort on the ground
to make this happen and the competition over post-Genocide Shingal continued thereafter.
A Survey of the Political History of the First Year of the Yezidi Genocide
27
X. The Origin of Hashd al-Shaʿbi Involvement
with the Yezidis
Not long after US officials had yet again made overtures to the Yezidis about
the idea of an ‘Iraqi national guard program that could accommodate localised
ethnoreligious components’ and then failed to pursue its development, an alternative that essentially mirrored this proposal began to take shape in Iraq.
In early November of 2014, three men made a decisive visit to Baghdad. Two
were Yezidis from Shingal — Murad Sharro of Gohbal and Jameel Chomer of
Borek — and the third was a well-connected Gergeri associate of Murad’s who
had agreed to facilitate some vital introductions. The Gergeri advised them not
to ‘waste time’ approaching Salih al-Mutlaq (then deputy prime minister) or
Osama al-Nujaifi (then vice president of Iraq and brother of the governor of
Nineveh) who, he maintained, would do nothing for the Yezidis, but recommended to instead focus on Shi’i resources that were emerging as powerful
players in the war with the Salafi jihadists.
In Baghdad, the three men met with Hadi al-ʿAmiri and Abu Mahdi al-Muhandis
and made a plea for direct Hashd al-Shaʿbi support for Yezidi fighters currently
defending Shingal against IS. The Hashd leaders found the proposal agreeable
and al-Muhandis stated that he would authorise support for five hundred fighters.
The Gergeri advocated for a higher number, citing the courage and commitment
of the fighters, and convinced al-Muhandis to support one thousand.
The next step was to draw up a list of fighters who would be the beneficiaries
of this support, a task that fell to the two Yezidis. Jameel suggested that they
request to be taken to Shingal by helicopter where they could visit each enclave
of resistance, from Piri Awra in the east to Jifriya in the west, and record the
name of every active fighter. Murad vetoed this idea and insisted on working
with his cousin Dawud Jindi (who at that time was deputy chief of the PUK’s
party branch for Nineveh) and their associate Haydar Shesho to formulate the
list, asserting that they would also be best positioned in Shingal to distribute the
funds to the fighters.
When Dawud and Haydar sent the draft list of one thousand names, Jameel became concerned, noting that some of the names belonged to relatives of
Dawud and Murad who had not been fighting but had only recently returned
to Shingal after the corridor had opened. When he confronted Murad, the latter contended that they were men who had pledged to soon join the resistance.
Jameel insisted on replacing some of the names with others whom he knew were
actively fighting in different areas of the mountain. The finalised list was then
submitted to Hashd officials.
The first Hashd al-Shaʿbi disbursement for the Yezidi fighters was transacted at
the end of that same month. The amount was calculated to provide a three-month
instalment of a monthly salary of 875,000 IQD delivered as a lump sum to each
28
Matthew Travis Barber
fighter on the list to cover the previous, current, and subsequent month (Oct.–Dec.).
Three months each for one thousand fighters at this rate totaled 2,625,000,000 IQD
(at the time equal to $2,187,500) that was dispensed in this first payment.
Had the salaries been distributed as arranged, they would have empowered
and expanded the core of displaced families tenaciously clinging to existence in
Shingal, since the remnant of Yezidis who refused to migrate to the IDP camps
of Dohuk was largely made up of the relatives of the early active fighters. But
little of the funding was to reach the fighters.
When the funds were first released, Murad Sharro transported them (as bundles of cash in duffle bags) into the ḥawāla44 office unaccompanied; no accountability measure was followed to ensure that all of the funds were transferred to
the intended recipient. It soon became clear that Murad had redirected funds to
himself, probably by transferring a portion to a family member while inside the
ḥawāla office. Compounding matters, whatever portion of the funds reached
Haydar Shesho were not distributed as had been agreed. In the first half of December, Jameel Chomer made calls to a number of fighters whose names were
on the list to confirm that they had received their salaries, but many reported
having received nothing. He assured them that Dawud Jindi would give them
their salaries and encouraged them to travel to Sharfeddin to see him. Fighters
coming from the north-western and southern parts of the mountain reported
back to Jameel of having been rebuffed by Jindi who told them “I’ve never seen
you and I don’t know you”. It soon became clear that the only fighters receiving
any money were part of Haydar’s circle within the Jiwana tribe. Even these were
receiving unequal amounts with no discernible consistency. It also appeared
that probably more than a million dollars were missing.
At this point, Jameel contacted Murad (who had remained in Baghdad) and
confronted him about what was happening. After Murad was only able to make
excuses, Jameel published on Facebook (in December) the details of the agreement with the Hashd al-Shaʿbi, a breakdown of the funding, and disclosed that
most fighters on the list were unpaid.
Murad Sharro became an instant pariah in the community and was unable
to return to the KRI. He remained in Baghdad from this point forward, purchasing a restaurant/night club that he operated thereafter. The ‘Hashd Body’
(Hayʾa al-Ḥashd) in Baghdad did not cut ties with Murad and there is speculation that some Hashd personnel may have colluded with him to receive some
portion of the stolen funds. Importantly, Haydar Shesho, now the main beneficiary of the funds that had reached Shingal, made no comment on Jameel’s
public revelation but remained silent regarding Murad’s actions for well over a
year. Haydar continued to receive Hashd funding in early 2015, a fact that he
confirmed in TV appearances. People in the community remarked on a sudden
44 A system of informal money transfer based on promissory notes, common in the Middle
East, which can function even in the absence of traditional banks.
A Survey of the Political History of the First Year of the Yezidi Genocide
29
augmentation of visible indicators of the material status of Haydar’s deputies.
Two hundred sets of full military uniforms and boots reserved for Yezidis by
the Hashd Body and shipped from Baghdad (over which one of Haydar’s deputies had responsibility) disappeared in Slemani and never reached Shingal. An
office for the ‘Hashd al-Shaʿbi of Sinjar’ was opened in Baghdad and staffed by
Murad Sharro and Izzet Kamo, the brother of Haydar Shesho’s wife. These two
collected Hashd al-Shaʿbi salaries but seemed to serve little function beyond
soliciting further financial donations from individuals for Yezidi resistance and
coordinating subsequent Hashd disbursements that were sent to Shesho.
The tragic results of the saga described here cannot be overstated. Used wisely
and responsibly, the initial support of the Hashd Body could have shaped an outcome in Shingal entirely dissimilar from the political morass that developed over
the following five years. Hashd funding (and the governmental support and administrative legitimacy conferred by it) should have laid the foundation for the
creation of non-partisan, local Yezidi security consistent with the communitywide demand that had reverberated from 4 August onward. During the phase of
initial resistance, the focus was entirely on fighting IS with no thought given to
politics. The heads of most fighter groups that had coalesced around the mountain remained unaffiliated and the inclusive use of Hashd support would have
created a non-partisan umbrella empowering fighters to remain in Shingal and
sustain an effective campaign of resistance apart from the KDP and not dependent on the Peshmerga. Haydar Shesho also had the unique opportunity to position himself as the head of such an umbrella, a direction that would have sustained
the early veneer of heroism that he enjoyed and cemented his authority in the eyes
of a community desperate for an uncompromising, apolitical leadership figure.
Instead, the squandering of this opportunity destroyed this potential trajectory. Though Hashd al-Shaʿbi support for the Yezidis had originally been conceived of as nonpartisan, Haydar became its direct beneficiary and it was this
patronage that enabled him to develop his HPŞ entity, which later evolved into
a political faction. More problematically, as Haydar was not active throughout
Shingal, the impact of early Hashd support for the Yezidis was for the most
part limited to the Sharfeddin area. Though Sharfeddin was a crucial front, this
emergent partiality functioned to the detriment of resistance in other parts
of the mountain and left the PKK as the only available patron for many other
pockets of hitherto unsupported fighters.
Above all, it was astounding that at the most desperate hour the Yezidis had
faced in their modern history, the welfare of the entire community was readily exchanged for the financial gain of a few individuals. Nevertheless, much of
the community was willing at this stage to turn a blind eye to hints of scandal.
Not only was corruption viewed as an omnipresent, inevitable, and normalised
feature of Iraqi politics and society, the community was desperate for any leadership that might inspire hope. With the knack of a politician, Haydar built a renowned persona and skilfully generated notoriety through media interviews. Up
30
Matthew Travis Barber
until the spring of 2015, therefore, the HPŞ attained more public visibility than
the YBŞ, despite being inferior to it in military efficacy and geographical spread.
The direct involvement of the Hashd al-Shaʿbi with the Yezidis established an
important precedent for the later expansion of the Hashd system among other
non-Shi’i communities in Iraq, including Assyrians and Sunni Arabs.
XI. The Liberation of the North Side of Shingal
IS enslaved more than 3,500 Yezidi women and girls on the first day of the Genocide. The jihadists began immediately distributing them throughout IS-held
territories in Iraq and Syria, but it took time to complete this process. Women
and girls were transferred out in stages and in some cases were moved back and
forth multiple times between the Tal ʿAfar, Shingal, and Baʿaj areas. Groups of
hundreds remained held in Yezidi areas of Shingal for months. Following the
Kocho Massacre, the empty village was used as a holding site for hundreds of
captives who were moved into it from other areas. Hundreds were also held in
various locations inside and around Shingal City.
From August through November, Yezidis transmitted detailed intelligence on
the enslavement situation to the US Department of Defense. Many of the captured women had kept their cell-phones hidden on their persons and were able
to call out to their relatives who would then connect them to Yezidis leading the
intelligence-gathering process. These volunteers were therefore able to construct
a detailed picture of the locations where women were held (sometimes down to
specific buildings identifiable in satellite images), numbers of women at those
locations, and numbers of jihadists guarding them. The number of IS sentinels
responsible for managing captives was in some locations so small that girls and
women were sometimes able to sneak away and escape on their own. This was
especially the case when coalition airstrikes would target nearby IS positions;
women were reporting that their captors would often flee and take cover leaving
them temporarily unguarded. When it became understood how feasible it could
be to perform several large-scale rescues, a small flame of hope was kindled.
The potential liberation of Shingal was therefore marked by an exceptional
urgency that differentiated it from the strategic need to reclaim other territories
that IS had seized. It was directly linked to a unique human element — a critical
opportunity to prevent hundreds of still-rescuable women and girls from experiencing perhaps the most horrible aspect of the larger Genocide project. For the
remainder of summer and then the entire autumn, the whole community waited
with attention fixed on the prospect of liberation. But as the months passed
without any visible sign that the KRG was prioritising a liberation operation,
the lack of action not only drew the censure of Yezidi leaders but also resulted in
A Survey of the Political History of the First Year of the Yezidi Genocide
31
widespread demonstrations in the IDP camps. These were cracked down upon
violently by secret police and some demonstrators were hospitalised.
As it became clear that the liberation of Shingal might not materialise, those
involved in the advocacy effort begged for US Special Forces extraction operations that could have rescued over one thousand women and girls who had not
yet disappeared into the sexual slavery machine, warning that the window of
opportunity to perform these rescues was closing quickly.
The Obama administration refused to order these operations. As the weeks
dragged on, Yezidi and non-Yezidi advocates engaging the US government
watched the window of opportunity close as IS gradually relocated women to
Syria. The window was entirely shut on 25 November, when an estimated 757
Yezidis being held at Kocho were transferred to Tal ʿAfar.45 These included many
whole families who had undergone the forced-conversion process; unmarried girls
were also separated from their families on the twenty-fifth and taken away. Two
days earlier, Peshmerga and Iraqi forces had worked together to retake Jalawla and
al-Saʿdiya, two towns of strategic value in Diyala Governorate, from IS. Watching
the US fail to act during the months leading up to this point was like witnessing
a slow-motion repeat of the Kocho Massacre — this time with foreknowledge of
potential outcomes and detailed awareness of the situation on the ground. The
men of this group of captives, along with others, were later massacred in Tal ʿAfar
after IS determined that their conversions to Islam were not genuine.
After months of waiting, the US-assisted Peshmerga operation to liberate
Shingal finally came three weeks following the closure of the rescue window; its
belated arrival, therefore, was as heart-breaking as it was welcome. Supported
by US airstrikes, the operation to retake the areas north of Shingal Mountain
was conducted on 17–22 December and ended the IS siege of the mountain.
KDP Peshmerga sent forces southwest from both Zummar and Rabiʿa; PKK/
YPG forces also participated. The community was astounded and further disappointed when the operation did not continue to the south side, leaving half
of the Shingal region under IS control; neither were villages west of Tal ʿAfar —
which contained large numbers of captive Yezidis — liberated.
This liberation inaugurated the next chapter in the political history of the
Genocide, not only by opening the route between Dohuk and Shingal that allowed civilians from the north side to access their homes, but also by re-establishing an extensive KDP presence for the first time since 3 August. This
substantialised the rivalry between the PKK, whose presence was now firmly
established and impossible to dislodge, and the KDP.
45
This count was performed by the Sinjar Crisis Management Team, a network of volunteers
headed by Murad Ismael that functioned primarily in 2014. Members of SCMT in Iraq,
initially led by Nareen Shammo, would gather data and convey it to members in the US
who would then generate intelligence reports that were passed to US officials.
32
Matthew Travis Barber
XII. The Shift of Hashd al-Shaʿbi Support
from the HPŞ to the YBŞ
The KDP’s no-tolerance policy regarding Yezidis of disputed territories who
politically aligned with Baghdad made the ‘arrest’ of Haydar Shesho on 5 April
2015 somewhat unsurprising. Haydar was seized by secret police while visiting
the Yezidi town of Khanke during his first trip to the KRI since his relationship
with the Hashd al-Shaʿbi had begun. At the same time, Haydar’s detainment
was startling to those who expected that his German citizenship,46 PUK membership, status as a former Iraqi MP, or role heading a high-profile militia in
the fight against IS would have protected him. It did seem rather brazen for the
KDP to detain a figure who led one segment of the grassroots defensive forces
that formed to fill the gap created by the Peshmerga withdrawal; nevertheless,
this arrest came after the liberation of the north side when the reassertion of
dominance became the KDP’s priority.
It is not difficult to see why the KDP viewed the HPŞ as a threat. In a public
speech at a conference in Germany in January 2015, Haydar had asserted that
Yezidis had a right to self-determination and had charged the Peshmerga with
responsibility for the Genocide: “No one should claim that they [the Peshmerga]
were not responsible for the Shingal region and the massacre. I consider the eight
thousand Peshmerga [in Shingal] and the three thousand in the Zummar region,
a total of eleven thousand Peshmerga, to be responsible for this genocide”.47 The
HPŞ inspired many Yezidis who hoped for the emergence of a viable ‘third way’
between the KDP and the PKK in Shingal. Haydar had not renounced his PUK
membership even after securing Hashd al-Shaʿbi sponsorship, but his status with
the party had become ambiguous. Haydar promoted the HPŞ as an independent
entity and the militia never flew PUK flags. Though the KDP would likely have
been more reticent to target a figure acting in the name of the PUK,48 Haydar’s
self-declared non-affiliation emboldened the KDP in their action against him.
46
47
48
Bild 2015.
EzidiPress 2015a.
It may be useful to further explain that in the years prior to the Genocide, many Yezidis
who disliked the KDP would join the PUK as an alternative. PUK membership usually
did not translate into real opportunities to meaningfully participate in governance and
administration in KDP-controlled areas; it did, however, offer a degree of protection
against KDP political violence that members of local, non-Kurdish parties did not have.
In the aftermath of the Genocide, the PUK was not looked to as a potential ‘third way’
between the KDP and the PKK as it was simply not a significant player in Shingal. Regarding the HPŞ, from the PUK perspective there was no reason to expel Haydar from
the party even though he was promoting his militia as an independent one. The token
PUK forces in Shingal notwithstanding, the party had no real claim to Shingal and the
HPŞ therefore posed no threat to party objectives. If anything, many in the PUK simply
ignored the rhetoric of non-affiliation and continued to claim Haydar as their own while
he was featured in the spotlight as a hero.
A Survey of the Political History of the First Year of the Yezidi Genocide
33
The arrest was distressing to the community. Yezidis in Germany held demonstrations and the German consulate in the KRI demanded that Haydar be
released.49 Accusations were once again flung at the KDP for ‘working on the
same side as Da’sh’. Haydar was initially jailed with IS members.50
That the arrest could not have been conducted in Shingal but was only possible inside the KRI illustrated an important dynamic that characterised Shingal
over the next several years. This was a kind of balance between factions that
emerged in the absence of any one group enjoying hegemony. Though various
militias predominated in certain locales, with spheres of influence sometimes
geographically overlapping, no single entity could rise above its peers to claim
absolute authority in Shingal. The KDP could not target its rivals with arrests or
political violence without risking reprisals or the outbreak of conflict. This created an environment in which a range of actors functioned with a great measure
of freedom and — despite their mutual competition — a surprising degree of
coexistence that would never have been tolerated inside the KRI.
Authority inside the KRI, however, was unambiguous. If the KDP was now unable to contain its political opponents in Shingal, it could still persecute them if they
ventured a visit to relatives in the KRI. Members of the YBŞ who came to the KRI
were thus targeted with intimidation and arrests early on; because of this, most of
its members remained inside Shingal for a year or even two without seeing loved
ones, unable to visit them a mere three-hour drive away out of fear for their safety.
Kurdish authorities gave ‘the creation of an illegal militia’ as the reason for
Haydar’s arrest.51 Perhaps ironically, its status in Baghdad as a registered Hashd
al-Shaʿbi entity gave the HPŞ more legality at the level of the central government
than was possessed by the various KDP militias.
Haydar was held for nine days until he agreed to end his relationship with the
Hashd al-Shaʿbi. Upon his release, Haydar publicly announced that he would
not maintain any militia that was not under the authority of the KRG’s Ministry of Peshmerga. For one year following this point, Haydar receded as a consequential player and the HPŞ became mostly dormant.
Despite the focus thus far on the political drama surrounding the HPŞ, the
YBŞ was from the beginning a more consequential actor militarily. Baghdad
was interested in maintaining its support for a local player that would compete with the KDP over the disputed territory and the severing of Haydar’s
ties with the Hashd al-Shaʿbi created an opportunity for a new relationship to
be built between Hashd authorities and the YBŞ. Negotiations for the Hashd
Body payment of YBŞ salaries were completed in July 2015 and the YBŞ became registered as a Hashd entity on 4 August. The shift of Hashd assistance to
the YBŞ enhanced the status of the latter. Baghdad’s sponsorship also favour49 Ekurd Daily 2015.
50 EzidiPress 2015c.
51 EzidiPress 2015b.
34
Matthew Travis Barber
ably impacted international perceptions of the YBŞ’s legitimacy amid Turkey’s
broad-brush accusations that PKK affiliates were terror groups.
Though it had used the term ‘Peshmerga’ to refer loosely to any resistance
element in Shingal whose loyalty it could claim in 2014, after it regained overland access to Shingal the KDP had worked in early 2015 to formalise a Yezidi
Peshmerga entity. This party militia was developed to compete with the YBŞ
and HPŞ. In this regard, the KDP had a major financial advantage over the HPŞ,
especially after bringing about the latter’s loss of Hashd al-Shaʿbi funding. By
providing salaries to fighters, the Yezidi Peshmerga was a powerful instrument
for purchasing loyalty. The sentiments produced by the Peshmerga withdrawal
were one thing; economic necessities were another. The Yezidi Peshmerga was a
kind of extension of the patronage system in that the militia’s leadership could
hand-pick recruits from loyalist families to be awarded an income. As the HPŞ
faded, the YBŞ and KDP entities would become the most relevant players from
mid-2015 onward.
XIII. The Tal ʿAfar Massacre
Many of the Yezidi families that underwent forced conversion in the first week
of the Genocide were later housed by IS in the vacated Shi’i neighbourhoods of
Tal ʿAfar. Here they were kept and monitored, the sincerity of their conversions
remaining in question. A string of successful escapes occurring over the course of
early 2015 resulted in a heavy cost carried by the majority that remained trapped.
After IS authorities became convinced that the conversions of this community were
not genuine, another round of summary executions of Yezidi males was performed.
Aspects of the Tal ʿAfar Massacre remain opaque, but the killings most likely
occurred between 29 April and 2 May 2015. On 28 April a group of Yezidi families successfully made their way out of the city but then became lost and had
to return to it, where they were apprehended. According to anonymous sources
inside Tal ʿAfar, eight men from these families were immediately executed when
recaptured. IS then rounded up the male52 Yezidis in the city and marched them
to a mosque where they were held as a group. One eyewitness reported seeing
a segment of the group separated out and driven in large trucks to a location
outside of the city, after which they were not seen again. Initial reports from
sources in the city (and subsequently reaffirmed by Iraqi officials) gave estimates ranging from three to six hundred killed. Yezidi officials were initially
reluctant to take a definite position based on these unverifiable reports, but on
26 April 2016, leaders in Lalish published a statement affirming that 410 Yezidi
52
The pattern in these cases was to execute adult and post-pubescent teen males; that was
probably the segment of the population targeted in this massacre, as well.
A Survey of the Political History of the First Year of the Yezidi Genocide
35
men in Tal ʿAfar had been missing for the past year.53 Following the liberation
of Tal ʿAfar in 2017, scores of bodies were discovered in several mass grave sites
outside of the city.54 Occurring a full nine months from the inception of the
Genocide, the Tal ʿAfar Massacre represented the continued results of the initial
failure to pursue a decisive course of action in August 2014.
XIV. Conclusion
This chapter has presented the beginning of a much longer story. The episodes
and topics briefly detailed here represent important components of the political
history of the Yezidi Genocide during its early phase. Each of these events or
themes deserves fuller treatment, as do other developments of this period that
could not be included here.
Though concise, this treatment of the subject should clearly demonstrate
that one of the most tragic aspects of the Genocide was that it was preventable.
Though the KDP was fully accountable for its pre-Genocide irresponsibility
as well as for the Peshmerga withdrawal, the US also bore some measure of
culpability for the negligence that resulted in Yezidi defencelessness on 3 August. Once the Genocide began, political strategies also occasioned a sluggish
US response. Thinking that the jihadist menace would constitute a powerful
wake-up call that would shake up the self-interested order in Baghdad, the US
opted for an overly restrained reaction to the crisis. The administration wanted
Maliki to learn his lesson, but that approach was a gamble: the price for that
policy was the chaos that engulfed the country. US airstrikes and humanitarian
airdrops saved thousands of lives during the first year of the Genocide; this military response, however, appeared to represent the absolute minimum of what
needed to be done. The deficient response continued through the autumn and
its consequences resurfaced conspicuously in the Tal ʿAfar Massacre. Perhaps
this assessment does not do adequate justice to the strategic calculus of the US;
the purpose here, however, is not to empathise with US policy objectives but to
understand the impact of these decisions on the Yezidi community.
The factors that allowed the Genocide to occur — first negligence and then
outright abandonment — shaped the orientation of the community toward local
political players in the aftermath and prompted Yezidis to seek alternatives to
the KDP Peshmerga for help in resisting IS. The refusal of the US to honour the
Yezidi request for direct involvement in brokering a new, nonpartisan, and inclusive security and administrative solution for Shingal resulted in different factions of Yezidis exploring divergent avenues of potential support. The ensuing
dynamics were marred by drama and corruption and the KDP reacted angrily
53 US State Department 2016.
54 Bor 2018, pp. 41–43.
36
Matthew Travis Barber
to the involvement of Baghdad. The absence of requested international leadership in a context of absolute political dysfunction paved the way for entrenched
political competition in the subsequent period.
The next phase of the Genocide’s political history became less about IS and
more about the political competition over Shingal. The situation from 2015 to
2017 would be defined by a fraught legacy of human rights violations and political abuses as the KDP worked to reassert hegemony over Shingal. This legacy,
more than the soon-to-diminish presence of the Islamic State, would become
the factor that most jeopardised the Yezidi future in Iraq.
Bibliography
Amnesty International 2014: Ethnic Cleansing on a Historic Scale: Islamic State’s Systematic Targeting of Minorities in Northern Iraq, 09.2014. London.
al-Arabia 2014: “Abu Muslim al-Turkmani: From Iraqi officer to slain ISIS deputy.”
19.12.2014.
Associated Press 2014: “Iraq air force to back Kurds fighting Islamists.” In: Online
Athens, 04.08.2014.
Barber, M. 2014a: “The Yazidis Are Not Getting Support.” In: Syria Comment,
05.10.2014.
— 2014b: “The Yazidis are starving, traumatised, and still unsafe.” In: New Statesman,
14.08.2014.
— 2016: “They That Remain: Syrian and Iraqi Christian Communities Amid the
Syria Conflict and the Rise of the Islamic State.” In: Hertzke, A. D./T.S. Shaw
(eds.): Christianity and Freedom, Volume 2: Contemporary Perspectives. New
York, pp. 453–488.
BBC 2014: “France launches first air strikes on IS in Iraq.” 19.09.2014.
Bild 2015: “Deutscher Jesiden-Kommandeur verhaftet.” 06.04.2015.
Bor, G. 2018: Working Against the Clock: Documenting Mass Graves of Yazidis
Killed by the Islamic State. Available Online: https://www.researchgate.net/
publication/326812184_Working_Against_the_Clock_Documenting_Mass_
Graves_of_Yazidis_Killed_by_the_Islamic_State [Last accessed 01.09.2021].
Bruinessen, M. van 1994: “Kurdish Nationalism and Competing Ethnic Loyalties”,
self-published English translation of “Nationalisme Kurde et Ethnicités Intra-Kurdes.” In: Peuples Méditerranéens no. 68–69.
Cetorelli, V./S. Ashraph 2019: A Demographic Documenation of ISIS’s Attack on
the Yazidi Village of Kocho. London.
Chulov, M./M. Rasool 2020: “Isis founding member confirmed by spies as group’s
new leader.” In: The Guardian, 20.01.2020.
Cooper, H./M. D. Shear/R. Nordland 2014: “U.S. Pulls Back on Plans for a
Mountain Rescue.” In: New York Times, 14.08.2014.
Ekurd Daily 2015: “Kurdistan president: No independent Yazidi unit nor flag will be
accepted.” 09.04.2015.
A Survey of the Political History of the First Year of the Yezidi Genocide
37
EzidiPress 2014: “Yezidi children and women massacred by Islamic terrorists.”
29.05.2014.
EzidiPress 2015a: “HPŞ-Oberkommandeur Heydar Shesho: ‘Êzîden haben das Recht
auf Selbstverwaltung’.” 14.01.2015.
EzidiPress 2015b: “Kurdish government: Shesho arrested by official order.” 06.04.2015.
EzidiPress 2015c: “Interview mit Heydar Shesho: ‘Unsere Einheit und unsere Flagge
sind unverhandelbar’.” 14.04.2015.
Gordon, M.R./E. Schmitt 2014: “U.S. Said to Rebuff Iraqi Request to Strike Militants.” In: New York Times, 11.06.2014.
Hanna, R./M. Barber 2017: Erasing Assyrians: How the KRG Abuses Human Rights,
Undermines Democracy, and Conquers Minority Homelands. Available online:
https://www.atour.com/news/assyria/20170925a.html [Last accessed 01.09.2021].
Hauslohner, A. 2014: “Shiite villagers describe ‘massacre’ in northern Iraq.” In:
Washington Post, 23.06.2014.
Heras, N. A. 2015: “The Struggle for Syria’s al-Hasakah Governorate: Kurds, the Islamic State and the IRGC.” In: Jamestown Foundation’s Terrorism Monitor 13:7,
04.2015, pp 8–11.
Homeland Security News Wire 2014: “Maliki says he will not step down to facilitate
U.S. air strikes against ISIS.” 19.06.2014.
Human Rights Watch 2014a: “Iraq: ISIS Kidnaps Shia Turkmen, Destroys Shrines.”
27.06.2014.
Human Rights Watch 2014b: “Iraq: ISIS Abducting, Killing, Expelling Minorities.”
19.07.2014.
Human Rights Watch 2014c: “Iraq: ISIS Executed Hundreds of Prison Inmates.”
30.10.2014.
Kikoler, N. 2015: “Our Generation is Gone:” The Islamic State’s Targeting of Iraqi
Minorities in Ninewa. Washington, D.C.
Murad, N. 2017: The Last Girl. New York.
Nordland, R. 2014: “Despite U.S. Claims, Yazidis Say Crisis Is Not Over”, In: New
York Times, 14.08.2014.
Obama White House Archive 2014 a: “Remarks by the President on the Situation in
Iraq.” 19.06.2014.
Obama White House Archive 2014 b: “Statement by the President on Iraq.” 11.08.2014.
Rovera, D. 2014: “Testimonies from Kocho: The village ISIS tried to wipe off the
map.” In: LSE Middle East Centre reports. London, 18.08.2014.
Shand, S. 2018: Sinjar: 14 Days That Saved the Yazidis from Islamic State. Lanham.
al-Tamimi, A. J. 2016: “An Account of Abu Bakr al-Baghdadi & Islamic State Succession Lines.” In: Pundicity, 24.01.2016.
— 2018: “The Biography of Abu Ali al-Anbari: Full Translation and Analysis.” In:
Pundicity, 17.12.2018.
— 2019: “Opposition to Abu Bakr al-Baghdadi: The Testimony of a Former Amni.”
In: Pundicity, 21.05.2019.
Toorn, C. van den 2014: “How the U.S.-favored Kurds Abandoned the Yazidis
when ISIS Attacked.” In: Daily Beast, 17.08.2014.
United Nations 2015: Report of the Office of the United Nations High Commissioner
for Human Rights on the human rights situation in Iraq in the light of abuses com-
38
Matthew Travis Barber
mitted by the so-called Islamic State in Iraq and the Levant and associated groups,
UN Doc. A/HRC/28/18, 27.03.2015.
United Nations 2020: Letter dated 20 January 2020 from the Chair of the Security
Council Committee pursuant to resolutions 1267 (1999), 1989 (2011) and 2253
(2015) concerning Islamic State in Iraq and the Levant (Da’esh), Al-Qaida and
associated individuals, groups, undertakings and entities addressed to the President of the Security Council, UN Doc. S/2020/53, 20.01.2020.
U.S. State Department 2016: 2016 Country Reports on Human Rights Practices: Iraq.
Whiteside, C. 2017: “A Pedigree of Terror: The Myth of the Ba’athist Influence in the
Islamic State Movement.” In: Perspectives on Terrorism 11:3, pp 2–18.
Yazda 2016: Mass Graves of Yazidis Killed by the Islamic State Organization or Local
Affiliates On or After August 3, 2014, 28.01.2016.
Zengerle, P./M. Spetalnick 2014: “In Washington, growing chorus calls for Iraq’s
Maliki to go.” In: Reuters, 18.06.2014.