Afghanistan Analysts Network – English

Context and Culture

From ‘Slavers’ to ‘Warlords’: Descriptions of Afghanistan’s Uzbeks in western writing

Christian Bleuer 26 min

From the early 1800s to the present day, western writers have explored Afghanistan either in person or from a distance, their publications providing a view of Afghanistan’s governments and people to the wider audience in Europe, the United States and the west. However, this view is distorted in many ways. One noticeable case in this regard is the analyses and mentions of Afghanistan’s ethnic Uzbek community. AAN guest analyst Christian Bleuer here explores the history of western writing on Uzbeks, from openly racist 19th century historians to 21st century journalists, with their more subtle anti-Uzbek overtones. He traces the portrayal of Uzbeks from the rise of the Afghan state to the recent elections – which produced the highest-ranking Uzbek in modern Afghan history, Abdul Rashid Dostum, now Afghanistan’s First Vice President. 

Uzbek emissary Mirza Faiz (left), representing the northern Uzbek Khan Mir Wali, meets with a representative of Afghan ruler Shah Shuja in September 1840. The lithograph by James Rattray is via the British Library

Often times, the way different ethnic communities in Afghanistan have been portrayed has been in service of certain political goals – of empire (pro-British Empire), ideology (anti- or pro-communism) or advocacy (pro- or anti-US government). At other times, particularly more recently, distorted or incomplete analyses are produced for convoluted and complex reasons. The reasons for the often insulting, biased or dismissive portrayals of Uzbeks in western publications and journalism have changed over time, but some factors remain constant and can be seen in both the writings of the British Empire and in more recent publications.

The age of imperial writing

The historian BD Hopkins in 2008 detailed how not only British interventions in Afghanistan during the imperial period were strongly ethnicised, but also British writing. Both prioritised Pashtuns while dismissing other ethnic groups. This prioritisation has continued through the various foreign interventions in Afghanistan up until the modern day, crossing over to others such as the Americans. From the influential early 19th century writings of Scottish historian and British India administrator Mountstuart Elphinstone (1) onward, Hopkins noted, the western – especially British – focus on Pashtuns

…narrowed British understandings of Afghan political society to one almost solely constituted by the Pashtun. […] While this privileging of the Pashtun is understandable in historical origin – Elphinstone did visit the court of the Durrani monarch – it has led to an underlying presumption political power ultimately rests with the Pashtun. The British were at pains to stress the tribal and genealogical legitimacy of those they supported, either directly or indirectly. They conceived an Afghan political entity where none but the Pashtun could be the ultimate arbiters, if not exercisers of power. This ‘Pashtunization’ continues to hold sway over both foreign interventions and popular conceptions of political authority in Afghanistan. (2)

The later 19th century works of British Surgeon-Major Henry Walter Bellew are equally as neglectful of the non-Pashtun people of Afghanistan as are the earlier publications. Bellew’s ethnographies of Afghanistan, for example his book The Races of Afghanistan, Being a Brief Account of the Principal Nations Inhabiting that Country, completely ignores Uzbeks and other Turkic people. In this book Bellew states that Uzbeks “exercise little, if any, influence in the affairs of the country as a whole, and need not now engage our attention.” (3) In his other works, the Uzbeks are merely acknowledged as existing, and a note is inserted saying they will be analysed at a later date in another publication, which was to be never forthcoming. (4)

Popular histories, travelogues and narratives of Afghanistan, which are often far more influential than their scholarly equivalents, have paid little attention to Afghan Uzbeks. When the Uzbeks do make an appearance in the British literature, the reference is usually dismissive or insulting. For example, the Scottish Naval Lieutenant John Wood wrote in his classic 1841 travelogue:

…the wisdom of a people is to be estimated by the length of their beards, but little of Solomon’s mental superiority is inherited by the Uzbeks.

The appearance of Kunduz accords with the habits of an Uzbek; and by its meanness, poverty, and filth, may be estimated the moral worth of its inhabitants. (5)

While John Wood may not have been a prominent figure in Afghanistan or British India, other writers with higher stature in the region made similar assessments of Uzbeks. The influential architect of Afghan government policy, Major Charles Edward Yate, commented while investigating northern Afghanistan in the 1880s:

The Usbegs [Uzbeks] certainly are neither a handsome nor an interesting race, so far as one can judge of those we see here.

The Usbegs of this country, I must say, do not strike me as a pleasant race. In [northwestern Afghanistan] the majority of the villagers appear a dirty, sullen-looking, lazy sort of people… (6)

The main reason for such anti-Uzbek commentary at around this time (1880s to 1890s) appears to have been the justification it implicitly gave to British support for the resettlement of large numbers of Pashtuns in northern Afghanistan for strategic reasons. British views of Uzbeks were heavily influenced by Afghan government policies meant to subdue and dilute the population of ethnic groups in the north, where the Durrani Pashtun rulers had previously conquered the now semi-independent Uzbek khanates. The British government subsidised and provided arms to Amir Abdul Rahman Khan in support of his campaign in the north of Afghanistan (1880s and 1890s). This campaign, which included a component generally referred to as ‘Pashtunisation’ or ‘Afghanisation’ (used synonymously in the literature and in Afghanistan in the sense of ‘Afghan’ meaning ‘Pashtun’) was, depending on the various historical interpretations, either an extended period of ethnic cleansing or merely a resettlement programme that supported Pashtun settlers in northern Afghanistan. For the Afghan government, it was about controlling disobedient non-Pashtun areas and exiling rebellious southern Pashtuns; for the British, it was about keeping the north of Afghanistan free of Russian influence. Regarding the strategic use of Pashtunisation, Colonel Yate, who coined the term ‘Afghanisation’ and was its architect while he was still a Major, wrote in 1893 that

It is only the non-Afghan tribes such as the Maimanah Uzbegs [Uzbeks], the Herati Hazarahs and Jamshidis, etc. that have any intercourse or communication with the Turkomans or Russians, and once encircled by Afghans they are safe.  (7)

According to the euphemistic assessment of British officer Captain PJ Maitland, a contributor to the Afghan Boundary Commission – whose mission it was in the late 1880s to define Afghanistan’s northwest borders – and a colleague of Yate’s, Abdul Rahman’s resettlement policies were aimed partly at thinning “the Usbak [Uzbek] population of Turkistan with people on whose support he can rely in case of foreign invasion…” (8) The British were allies and co-architects with him in this regard. The ‘Yate plan’ was supported with the goal of having the Pashtuns dominate the political, social and agrarian life of Afghan Turkestan. (9) Maitland, through the Gazetteer of Afghan Turkestan, remarked

This immigration is encouraged by the Amir for obvious reasons. There is plenty of room for a much larger population than now exists and it is possible that if the province remains Afghan, and at peace, the Turki-speaking population [eg, Uzbeks] may come to be a minority in the next 20 or 30 years. (10)

The historian JL Lee, writing in 1996, suspects that a possibly deliberate undercount of the north served the purposes of both the British and Amir Abdul Rahman. They wanted to represent the north as under-populated to justify its Pashtunisation while the Amir wanted to suppress the acknowledged number of Uzbeks, Turkmen, Hazaras and Tajiks while boosting the estimates for the numerical superiority of Pashtuns in Afghanistan. Lee makes the comparison of Abdul Rahman’s and Maitland’s representation of the north as an underpopulated ‘wasteland’ to Theodor Herzl’s writings that sought to justify the colonisation of Palestine: “This ‘wasteland’ theory completely disregarded the rights, wishes, or indeed, the existence, of the third of a million or so native inhabitants of Lesser Turkistan [northern Afghanistan]…”  (11)

Wood was writing in 1841 while Yate wrote his book in 1888. The reasons for these prejudiced attitudes, here and in many other publications that span this era, are in many ways unclear and open to debate. The drivers for anti-Uzbek sentiments in historical literature, both scholarly and popular, include, but are not limited to, the following:

a) British support for a Pashtun monarchy in opposition to northern ethnic groups that were viewed as too close to Russian-controlled Central Asia (until the late 1800s) (12);

b) The Mongol and Turk invasions creating long-lasting anti-Turk/Mongol historical memories in Europe that were later transferred to Turkic Uzbeks (18 and 19th century) (13);

c) The ‘gatekeeper effect’ whereby visitors to Afghanistan first arrive in Kabul from the direction of British India (or occasionally Persia) and usually first meet only Pashtuns and Tajiks who then get to describe their Uzbek and Hazara compatriots, often in negative ways (18th century to well after the mid-20th century);

d) The popularity of racial theories of Aryan superiority (19th century to mid-20th century), wherein ‘Aryan’ Indo-European Pashtuns were compared favourably to the ‘Mongol’ Turkic Uzbeks (14);

e) The influence of the popular ‘Israelite origin’ theory in Europe (19th century) wherein Pashtuns are described favourably as a ‘Lost Tribe of Israel’ (15);

f) The terrible treatment that the Uzbek rulers of the Bukharan Emirate gave to not just their own citizens, but especially to the British envoys Stoddart and Conolly, as well as to other British travellers and spies (ie, extended torture followed by decapitation).(16)

Non-British writing on Afghanistan’s Uzbeks

Explicitly racist descriptions that single out Uzbeks for abuse amongst Afghanistan’s diverse population are not the exclusion domain of British military writers, nor of British writers in general. A prominent historical example here is provided by the first American to visit Afghanistan. Josiah Harlan was a Quaker turned con-artist who enlisted as a surgeon in the British East India Company despite his lack of medical education. He would later work his way into Afghan Amir Dost Muhammad Khan’s service and lead an Afghan attack in 1838 on the powerful northern Uzbek khan, Murad Beg. Regarding the Uzbeks, Harlan in his memoirs labelled them as cruel, cowardly slavers and “indolent scoundrels.” (17) Most British writers of the 19th century focused their anti-slavery rhetoric on Uzbeks and Turkmens while downplaying or neglecting to mention Pashtun enslavement of Hazaras, Kafirs [Nuristanis] and others, including Uzbeks themselves. The abolitionist Harlan actually does mention Pashtun participation in slavery, but he went far and beyond others in his comments about Uzbeks. These are just a small selection of his descriptions: (18)

In the practices of hospitality, so universally prevalent amongst the votaries of Islam, the Uzbecks are miserably deficient. 

The Uzbecks would rather sell [enslave] than feast [host] a traveller.

The Uzbeck is a merciless robber, who spares neither men, women or children. He is infamous for his cruelty, of which his southern neighbours say, “the mercy of an Uzbeck is equivalent to the anger of an Avghaun!”

The Uzbecks are an unsocial, cold hearted race…

Remarkable here is that German, Scandinavian and Russian historians and travellers who arrived in Afghanistan or Central Asia from the north or west did not carry the same level of virulent anti-Uzbek prejudice. (19) Some even took a special interest in Turkic peoples (eg, Turkmens, Uzbeks, Kyrgyz, etc), the later example of Swedish scholar Gunnar Jarring being particularly notable for coming to Afghanistan especially to research its Turkic population. (20). However, the volume and influence of publications by British writers overwhelmed those of other Europeans, both in scholarly and popular publications.

The post-imperial age of writing

Years of heavily politicised pseudo-scholarly inquiry eventually gave way to more reasonable analysis once the Afghan central government solidified its control over the Uzbeks and after the Russian ‘threat’ to British India had subsided. The need to denigrate the internal enemies (ie, the Uzbeks) of a valuable ally eventually disappeared. Notable here is the work of British Brigadier General Percy Sykes, whose 1940 publication A History of Afghanistan (21) is notable for both its neutral treatment of Uzbeks and its acknowledgement of the significant role they played in the history of Afghanistan and the empires that preceded it. Nevertheless, popular writings around this time gave little attention to Uzbeks – negative or positive. Mentions, though, were mostly dismissive, exemplified by Ernest Fox’s brief remark about the Uzbeks, along with others, in his 1943 Afghanistan travelogue: “The Hazaras, Uzbegs, Kafirs, and other tribes within the country are subject races.” (22) This was in line with Hopkins’ assessment or the earlier era: “later colonial travellers, administrators and ethnographers perpetuated this focus on the Pashtun, noting other ethnic groups largely for their subservience to Pashtun political authority.” (23)

Soon afterwards, anthropologists and others produced scholarly work based on fieldwork in Afghanistan from the late 1950s to late 1970s. They dispensed with racist descriptions of the Uzbeks. However, few scholars were interested in researching this community, and study of them remained sparse. This disinterest in ethnic Uzbeks as a subject of study is a complex phenomenon, and was no longer tied to any favouritism vis-à-vis the Pashtuns, but more to do with the fact that anthropologists have typically favoured people who live in mountains, jungles or islands, or nomadic communities, rather than easy to reach plains or towns (see also this previous AAN dispatch).

Despite the paucity of studies on Uzbeks in Afghanistan, as compared to Pashtuns, Hazaras and even Nuristanis, there are a few excellent ethnographic and historical sources to work from. (24) However, elsewhere the information is scattered throughout smaller references in books on subjects other than Uzbeks, (25) or hidden in hard to find academic literature. (26) From this, it is, however, possible to construct a fairer and more nuanced narrative of Uzbek history – and an analysis of Uzbek society – that can help to illuminate the role of the Uzbek community in Afghanistan’s years of conflict, from 1979 to the present [Note: an AAN report on Afghanistan’s Uzbeks is forthcoming].

Journalists and analysts framing of the Uzbeks of Afghanistan, 1979 to the present day

After almost a century of insult and invisibility, the decades of war that commenced in 1978 provided Afghanistan’s Uzbeks with a renewed opportunity to contribute to the course of history in Afghanistan. Despite the role that ethnic identity had begun to play during the jihad of the 1980s, western journalists and analysts would mention ethnicity as important, but their analysis was superficial and brief. Instead, the focus was on Islam as both an identity and a tool of mobilisation against the Afghan communist government. As a result, the references to ethnic Uzbeks during the 1980s are few and far between. (27) However, ethnicity became more fashionable among journalists, analysts and other writers in the post-communist era, influenced by events in the former Yugoslavia and in other ‘ethnic conflict’ zones. This leaked over into Afghanistan studies and reporting, and ethnicity was then given a deeper analysis. (28) Ethnic identity had been very relevant in the 1980s, but this was not analysed to any great degree in western publications, except by a few journalists and certain scholars, most of whom had been writing on Afghanistan in the previous decades. (29)

However, at the end of the period of analysts focusing their attention on ethnicity (1990s), and well after the period of racist history and travelogue books had subsided, we saw similar stereotyping re-emerging. The Pakistani journalist Ahmed Rashid, for example, who writes for western publications and publishers, wrote in his widely read and cited 2000 book Taliban:

The Uzbeks, the roughest and toughest of all the Central Asian nationalities, are noted for their love of marauding and pillaging – a hangover from their origins as part of Genghis Khan’s hordes…

Uzbek clan history is a long litany of blood feuds, revenge killings, power struggles, loot and plunder and disputes over women.

Mahmud ibn Wali, a 16th century historian, described the early Uzbeks as ‘famed for their bad nature, swiftness, audacity and boldness’ and revelling in their outlaw image. Little has changed in the Uzbek desire for power and influence since then. (30)

This sort of derogatory language directed at ethnic Uzbeks is rare in modern scholarship on Afghanistan. (31)  However, the two editions of Rashid’s book tower far above any other work on Afghanistan. The two editions of Taliban are cited by a combined 1700 books and articles, double the citations of Steve Coll’s Ghost Wars and almost triple the citations of Barnett Rubin’s The Fragmentation of Afghanistan, the second and third most widely cited books on Afghanistan. (32) No author has been able to frame the image of the Uzbeks with such a wide audience, before or since. While no other authors or analysts attack Uzbeks in such a direct manner, there remains a not-so-subtle bias against Uzbeks in both journalism and academia. For example, it is common in analysis by prominent media outlets on the immediate post-2001 period to focus on Hazara and Uzbek perpetrators of violence while neglecting to mention Tajik perpetrators. A good example here is reporting on ‘revenge’ attacks by Northern Alliance factions on Pashtun civilians in the north after the defeat of the Taleban in 2001. The source of most of the coverage, a Human Rights Watch report, clearly stated that the largely Uzbek Jombesh, Hazara Wahdat-e Melli and Tajik Jamiat-e Islami all perpetrated abuses, yet media reporting tended to mention only the Uzbek and Hazara perpetrators. (33) This type of reporting by omission has continued through to the present day, including in election coverage. (34)

Aside from such specific incidents, the language used by journalists and analysts to describe Uzbek and non-Uzbek commanders differs strongly. For example, the term ‘warlord’ is used in the popular non-scholarly writing on Afghanistan in the negative sense of an armed commander who has achieved political power and who has a history of human rights abuses. As noted by Antonio Giustozzi, in Afghanistan “everyone from international pundits to local governors uses the term to discredit certain political factions or insult the “bad guys.” (35) There is no shortage of ‘bad guys’ from across all ethnic groups in Afghanistan. Yet, assessing the usage of the terms in all English language books in the year of its peak usage, 2003, ‘Uzbek warlord’ was six times more prevalent than ‘Pashtun warlord’ or ‘Tajik warlord,’ with the usage of ‘Hazara warlord’ too low to be charted. (36)

Despite the large cast of powerful armed commanders in Afghanistan, Uzbeks and the Uzbek commander Abdul Rashid Dostum are singled out and characterised as warlords out of all proportion to their share of the ‘warlord’ ranks or participation in war crimes. For example, brief biographical details used in a 2014 BBC report that outlined the competing election teams were published time and again. (37) They listed members of Ashraf Ghani and Abdullah’s teams in neutral terms: Ata Muhammad Nur is labelled as the “wealthy Balkh governor,” Muhammad Mohaqeq as a “powerful leader of ethnic Hazaras,” Gul Agha Sherzai as an “influential Pashtun” and Ahmad Zia Massud as brother of a “famous resistance hero.” The BBC singles out Dostum, alone, for denigration: he is “an Uzbek ex-warlord accused of human rights abuses.” (38)

The usage of ‘warlord’ in the scholarly literature is more neutral, but still negative and usually used to describe “a military leader who has political power, but little or no political legitimacy, both internally and externally.” (39) There are numerous commanders and faction leaders who fit this description in Afghanistan, especially in the 1990s and the post-2001 years. But a statistical search strictly of the academic and scholarly literature shows a very strong tendency to label Uzbeks as warlords while neglecting to similarly describe Pashtun, Tajik and Hazara commanders and politicians, as such. (40) As for just the scholars and academics, they are actually more likely to describe Uzbeks as ‘warlords’ when compared to references across all media. (41)

Any argument that Rashid Dostum and Uzbeks are more often described as ‘warlords’ because of their particular history of human rights abuses is not based on a reading of the war crimes literature. The most prominent of the comprehensive published surveys of the war crimes of the 1978-2001 conflict (the UN Mapping report and reports by Human Rights Watch and the Afghanistan Justice Project) (42) do indeed describe numerous abuses and atrocities committed by Dostum’s faction. Yet in terms of numbers and severity of abuses committed, other factions and their leaders appear as frequently in these human rights reports. Indeed, when accountability for war crimes became a hot political topic, in 2005, an indication of how broad the past participation in human rights abuses had been could be seen in the number and variety of ethnic background of the powerful figures who lobbied, successfully, for an amnesty bill that granted immunity for all past (and some future) war crimes. (43) Yet scholars, journalists and analysts appear reluctant to name other prominent commanders and faction leaders as ‘warlords.’ Non-Uzbeks are much more likely to be described simply by their position in the post-2001 government, or as a ‘tribal leader,’ ‘commander,’ ‘politician,’ or ‘hero.’  (44)

The example of Dostum is particularly illuminating. Journalists writing in the post-2001 period in Afghanistan have used a variety of disparaging descriptions for him. Eventually they narrowed down to certain popular adjectives. The most notable here is the use of ‘notorious.’ Keyword searches ranked by date show that, beginning in late 2001 and early 2002 a variety of anti-war blogs and advocacy websites began using ‘notorious’ to describe Dostum, whether in original writing or in excerpts. (45) Only one large publication, the Los Angeles Times, uses ‘notorious.’ (46) However, ‘notorious’ is eventually adopted en masse – by the Telegraph, McClatchy, Washington Post, Radio Free Europe/Radio Liberty, Huffington Post, al-Jazeera, New York Times, Al Ahram, New Republic, and others when introducing Dostum. (47)

What is remarkable here is not just that journalists are clearly influenced – knowingly or unknowingly – by the language used by other writers, but that the websites first using ‘notorious’ were, with the exception of the Los Angeles Times, advocacy platforms that harshly criticised not just Dostum, but most armed actors and American-aligned commanders, as well as the Taleban. Notable here is RAWA, whose criticisms spare no armed commanders, Uzbek or non-Uzbek. Yet it was Dostum to whose reputation the disparaging adjectives stuck the easiest. This is just one small example of how Dostum is a magnet for criticism and critical assessments. (48)

So while journalists and analysts likely borrowed their language from some of the advocacy websites that were the quickest to provide analysis and commentary, it is clear they did not borrow their prejudices from these websites. If they did, these journalists and analysts would also be equally harsh in their characterisations of all commanders active in the post-2001 period. (49)

Rashid Dostum often grants interviews to journalists, but this usually does nothing to help his reputation, which clearly precedes him. For example, Magsie Hamilton-Little of The Telegraph writes of her meeting with Dostum, described as “Afghanistan’s most fearsome warlord”:

At six feet tall, he towers a good seven inches above me. I feel my legs wobble – his reputation is nearly as fearsome as his bushy moustache. He is General Abdul Rashid Dostum, Afghanistan’s most powerful warlord, head of the Uzbek tribe…

It is only when I tell myself I have nothing to fear that I remember the allegations that make your stomach turn and heart pound, such as the one about how, at Dostum’s command, women had been raped and their breasts cut off before they were killed during the siege of Kabul in 1991.[sic] […]

On the way down in the lift, the sense of apprehension I might previously have felt has all but evaporated as I realise I have survived the meeting unscathed. (50)

Dostum’s absolutely hopeless position vis-à-vis his reputation – online in English – is exemplified in how Google’s search function autocompletes suggestions after his name. The autocomplete search suggestions are a reflection of what people are searching, what is written about that person on the internet, and ‘relevance’. (51) A Google search autocompletes “war crimes” when Dostum’s name is entered. Nobody else gets that, not Gulbuddin Hekmatyar, former President Najibullah, Ata Mohammad Nur, Rassul Sayyaf, Gul Agha Sherzai or Mullah Omar. (52)

Sayyaf and Hekmatyar are occasionally called warlords, but their Pashtun identity is seldom mentioned at the same time when compared to Uzbek commanders who are also labelled as warlords. The same works for Ahmad Shah Massud and Ismail Khan and their Tajik identity; they are rarely identified as Tajiks when the term ‘warlord’ is used. For Dostum, his being a ‘warlord’ and an Uzbek appears to have become inseparable. This is especially true for many scholars and academics, none of whom explain why, in their analysis, Uzbek commanders are five to ten times more likely than Pashtuns or Tajiks to be identified by ethnicity when being termed ‘warlords’. (53)

Potential explanations

What accounts for journalists and analysts focusing their negative coverage on Rashid Dostum while neglecting equally or more powerful people such as Ata Mohammad Nur, Marshal Fahim, Muhammad Mohaqiq, or Rassul Sayyaf? And why do ethnic Uzbeks in general still get singled out for negative media coverage when compared to other ethnic groups in Afghanistan? The following explanations have been offered to this author: (54)

a) Fear of negative consequences, including physical harm, for criticising powerful Tajiks and Pashtuns who base themselves in Kabul, whether in the government or not. In contrast, Dostum does not induce the same level of fear in foreign and local journalists and analysts in Kabul, with the above-cited writer for The Telegraph being an example of a journalist overcoming their fear rather than self-censoring out of worry over repercussions;

b) In Kabul, as well as in western Europe, the United States, Canada and Australia, there is an informal lobby of well-educated ethnic Tajiks and Pashtuns who are in positions (in academia, the media, various advocacy groups, the government, etc) to influence foreign officials, analysts and journalists. Uzbeks are lacking in this regard;

c) The ethnic Uzbek community in Afghanistan has the lowest level of literacy in the country. (55) This sets them at a disadvantage by restricting their ability to communicate their views in academia, media, social media, etc. This factor, along with historical and social factors that steer them towards certain trades and professions, strongly reduces the likelihood that they will be qualified to work in international organisations, NGOs, foreign government missions and for the local and foreign media in comparison to ethnic Tajiks, Hazaras and Pashtuns;

d) Ethnic favouritism in the expatriate community. For example, foreigners, including journalists, could be worried about offending their Tajik, Pashtun or Hazara co-workers or employees by calling prominent leaders in their communities ‘warlords.’ By contrast, foreigners would be less likely to be working with Uzbeks for the reason given above;

e) The disproportionate Uzbek support for the Afghan Communist government and Rashid Dostum’s past as a pro-communist militia commander created ill-will among many western journalists, writers and advocates that persists and was transferred to a new generation of journalists and analysts. This contrasts to the praise and glorification of Ahmad Shah Massud and his forces that can easily be found in western journalism and literature, especially in the 1980s and 1990s. Rashid Dostum’s very few western defenders’ works are marginal and ignored (56);

f) The anti-Uzbek racism in the historical literature has filtered down indirectly to modern analysis and journalists. Ahmed Rashid’s work is the most prominent in this regard;

g) Racism in Afghan society against Uzbeks (57);

h) The distance from Kabul of the bulk of the ethnic Uzbek communities. The further a commander or politician is from Kabul and from a position of power in the central government, the more likely it is they will be called a warlord;

i) Of all the crimes against civilians and prisoners of war between 1992 and 2002, Dostum’s fighters’ killing of Taleban prisoners at Dasht-e Leili is the only large-scale incident that can plausibly be tied to American forces (for details of the accusations, see AAN reporting here). As a result, the desire to pin war crimes on American forces has caught Uzbek forces by association while leaving crimes committed by non-Uzbeks as a lower priority for many journalists, human rights activists (not including Human Rights Watch (58) or the Afghanistan Justice Project) and various advocates in the immediate post-2001 period.

j) The effectiveness of the central government’s campaign to side-line Rashid Dostum after 2001, both by President Karzai and by rival Tajik commanders and politicians;

k) Rashid Dostum’s lack of charisma (as interpreted by western journalists and analysts), his poor public relations skills, and his ill-advised behaviour, such as when he detained and personally beat his former ally Akbar Bai. (59)

200 hundred years of writing on ethnic Uzbeks

It remains to be seen whether or not the recent rise of Rashid Dostum to the office of First Vice President, the highest official government post ever held by an Uzbek, will do anything to reduce negative coverage of ethnic Uzbeks in Afghanistan, or to encourage positive coverage. For now it is still clear that by omission, accusation, labelling or insinuation, Uzbeks fare poorly in both the academic literature and in the media coverage. And it has become increasingly difficult to find any analysis or discussions on Afghanistan’s Uzbeks that does not prominently mention Rashid Dostum – as indeed, this very dispatch has done. This is understandable, given his status over the last three decades as the most powerful individual within the broader Uzbek community in Afghanistan. However, this association with Dostum is a liability for Uzbeks in regard to how their community is portrayed by western writers. Dostum’s poor reputation will very likely remain as it is. Journalists and analysts have displayed an ever increasing tendency to frame Dostum as a war crimes committing ‘warlord’ while demonstrating a tendency, by comparison, to overlook or actively omit the crimes committed by non-Uzbeks.

The result of the first round of the 2014 election showed that the Uzbek vote in Afghanistan was split (see AAN reporting here): Rashid Dostum does not speak for the Uzbeks of Afghanistan in the same way that he has never represented them as a whole. Ethnic Uzbeks deserve to be analysed and portrayed separately from Dostum. In the early 1800s, it was Murad Beg, the powerful northern Uzbek khan who stood in for the image of the Uzbek in western literature as the ‘cruel’ Uzbek ‘slaver.’ There was never an Uzbek as powerful as him until the rise of Rashid Dostum, who, out of Afghanistan’s approximately three million Uzbeks, receives by far the most attention. Asking a foreigner in Kabul to name Afghanistan’s second or third most powerful Uzbek will likely be met with non-answers.

The portrayal of Afghanistan’s Uzbeks in western literature has, however, improved greatly since the days when an entire ethnic group could be discounted as ‘dirty,’ ‘lazy,’ ‘cruel’ and mentally deficient. This sort of writing ceased once there was no need for the British Empire to denigrate a rival for power to their clients in Kabul. A neutral period of writing existed from the 1940s until the 1970s, driven more by academic curiosity than in the periods before and after. Yet now a new set of subtle biases against Uzbeks have taken root and persists, most surprisingly in peer-reviewed academic publications. But it is the journalists who remain the most influential in creating popular images, giving the western readership a clear view of the typical Afghan ‘warlord’: an Uzbek. Afghanistan’s Uzbeks likely have a long way to go before they are given a fairer portrayal in Western publications.


(1) Mountstuart Elphinstone, An account of the kingdom of Caubul and its dependencies in Persia, Tartary and India: Comprising a view of the Afghan nation and a history of the Dooraunee monarchy (London, Longman 1815).

(2) BD Hopkins, The Making of Modern Afghanistan (New York, Palgrave Macmillan 2008) 32.

(3) Henry Walter Bellew, The Races of Afghanistan, being a brief account of the principal nations inhabiting that country (London: Thacker, Spink & Co, 1880) 14. See his other works: Introductory remarks to an inquiry into the ethnography of Afghanistan (London, Butler & Tanner 1891); An enquiry into the ethnography of Afghanistan (Woking, UK, Oriental University Institute 1891).

(4) Gunnar Jarring, On the distribution of Turk tribes in Afghanistan (Lund/Leipzig: CWK Gleerup/Otto Harrasowitz, 1939) 10.

(5) Lt John Wood, A Personal Narrative of a Journey to the Source of the River Oxus: By the Route of the Indus, Kabul, and Badakhshan, Performed Under the Sanction of the Supreme Government of India, in the Years 1836, 1837, and 1838 (London, John Murray 1841) 214-5.

(6) Charles Edward Yate, Northern Afghanistan: or, Letters from the Afghan boundary commission (Edinburgh, W Blackwood & Sons, 1888) 130, 246.

(7) JL Lee, The Ancient Supremacy: Bukhara, Afghanistan and the Battle for Balkh, 1731-1901 (Leiden/New York, EJ Brill 1996) 483, 595.

(8) Nancy Tapper, ‘Abd Al-Rahman’s North-West Frontier: The Pashtun Colonisation of Afghan Turkistan’, in The Conflict of Tribe and State in Iran and Afghanistan, ed Richard Tapper (New York, St Martin’s Press 1983) 239.

(9) JL Lee, The Ancient Supremacy: Bukhara, Afghanistan and the Battle for Balkh, 1731-1901 (Leiden/New York, EJ Brill 1996) 482.

(10) JL Lee, The Ancient Supremacy: Bukhara, Afghanistan and the Battle for Balkh, 1731-1901 (Leiden/New York, EJ Brill 1996) 483.

(11) JL Lee, The Ancient Supremacy: Bukhara, Afghanistan and the Battle for Balkh, 1731-1901 (Leiden/New York, EJ Brill 1996) 482-3.

(12) See, for example: BD Hopkins, The Making of Modern Afghanistan (New York, Palgrave Macmillan 2008) 32; JL Lee, The Ancient Supremacy: Bukhara, Afghanistan and the Battle for Balkh, 1731-1901 (Leiden/New York, EJ Brill 1996) 482-3.

(13) On European images and historical imagination of the Turks, see: Mustafa Soykut, Historical Image of the Turk in Europe, 15th Century to the Present (Istanbul, The Isis Press, 2003).

(14) Prominent examples include: Joseph Arthur Comte de Gobineau, Essai sur l’inégalité des races humaines (Paris, Firmin Didot 1853) ; Joseph Pomeroy Widney, Race Life of the Aryan Peoples, Volumes 1 and 2 (New York, Funk & Wagnalls 1907); Georges Vacher de Lapouge, L’Aryen: Son Rôle Social (Paris, Albert Fontemoing 1899). For a modern academic article on ‘Aryans,’ see: R Schmitt, ‘Aryans’, Encyclopaedia Iranica

(15) See, for example: George Moore, The lost tribes and the Saxons of the East and the West (London, Longman, Green, Longman, and Roberts, 1861); JP Ferrier, History of the Afghans (London, J Murray, 1858).

(16) The fate of Stoddart and Conolly was a popular topic at the time. See, for example: Rev Joseph Wolff, Narrative of a mission to Bokhara, in the years 1843-1845, to ascertain the fate of Colonel Stoddart and Captain Conolly (London, JW Parker 1845).

(17) Josiah Harlan, Central Asia: Personal Narrative of General Josiah Harlan, 1823-1841, edited by Frank E Ross (London, Luzac & Co 1939) 61-2, 83.

(18) Josiah Harlan, Central Asia: Personal Narrative of General Josiah Harlan, 1823-1841, edited by Frank E Ross (London, Luzac & Co 1939) 69-70, 78.

(19) Those who travelled in Central Asia excluding Afghanistan were even more positively disposed towards Uzbeks. The famous explorer Sven Hedin, for example, travelled through Central Asia with an Uzbek guide and confidant who he held in the highest regard. George Kish, To the Heart of Asia: The Life of Sven Hedin (Ann Arbor, University of Michigan Press 1985).

(20) Gunnar Jarring, On the distribution of Turk tribes in Afghanistan (Lund/Leipzig: CWK Gleerup/Otto Harrasowitz, 1939).

(21) Percy Sykes, A History of Afghanistan, Vols 1&2 (London, MacMillan & Co, 1940).

(22) Ernest F Fox, Travels in Afghanistan, 1937-1938 (New York, MacMillan 1943) x.

(23) BD Hopkins, The Making of Modern Afghanistan (New York, Palgrave Macmillan 2008) 32.

(24) Notable here are publications by Audrey Shalinsky and Gabriel Rasuly-Paleczek: Audrey C. Shalinsky, Long Years of Exile: Central Asia Refugees in Afghanistan and Pakistan (New York, University Press of America 1994); Gabriele Rasuly-Paleczek, ‘Ethnic Identity versus Nationalism: The Uzbeks of North-Eastern Afghanistan and the Afghan State’, in Post-Soviet Central Asia, eds Touraj Atabaki and John O’Kane (London, Tauris 1998) 216; Gabriel Rasuly-Paleczek, ‘The Struggle for the Afghan State: Centralization, Nationalism and their Discontents,’ in Identity Politics in Central Asia and the Muslim World: Nationalism, Ethnicity and Labour in the Twentieth Century, eds Willem Van Schendel and Erik J. Zurcher (London, I.B. Tauris Publishers 2001).

(25) For example: Pierre Centlivres and Micheline Centlivres-Demont, ‘Tajikistan and Afghanistan: The Ethnic Groups on Either Side of the Border’, in Tajikistan: The Trials of Independence, eds Mohammad-Reza Djalili, Frederic Grare and Shirin Akiner (New York, St. Martin’s Press 1997).

(26) For example, a publication of the Polish Academy of Sciences: Marek Gawęcki, ‘Structure and Organization of the Rural Communities of Central and Northern Afghanistan’, Ethnologia Polona 12 (1986).

(27) Source: author’s own research on ethnic Uzbeks in Afghanistan for MA thesis, 2005-2007.

(28) Source: a comparison of publications in the 1980s and 1990s, as compiled in The Afghanistan Analyst Bibliography, 6th edition, 2011.

(29) Good, but rare, contemporaneous examples here are: Robert L Canfield, ‘Ethnic, Regional, and Sectarian Alignments in Afghanistan’, in The State, Religion, and Ethnic Politics: Afghanistan, Iran, and Pakistan, eds Ali Banuazizi and Myron Weiner (Syracuse, NY, Syracuse University Press 1986); Jan-Heeren Grevemeyer, ‘Religion, Ethnizität und Nationalismus im afghanischen Widerstand’, Leviathan, Vol 13, No 1, 1985, 115-128; Alfred Janata, ‘Afghanistan: the ethnic dimension’, in The Cultural Basis of Afghan Nationalism, eds E W Anderson and N Hatch Dupree,  Oxford University Press 1990; Eden Naby, ‘Ethnic Factors in Afghanistan’s Future’, in The Tragedy of Afghanistan: The Social, Cultural, and Political Impact of the Soviet Invasion, eds Bo Huldt and Erland Jansson, London: Croom Helm Limited 1988; Audrey C Shalinsky, ‘Ethnic reactions to the current regime in Afghanistan: A case study’, Central Asian Survey, Vol 3, No 4, 1984; Richard Tapper, ‘Ethnicity and Class: Dimensions of Conflict’, in Revolutions and Rebellions in Afghanistan: Anthropological Perspectives, eds M Nazif Shahrani and Robert L Canfield, University of California Berkeley 1984.

(30) Ahmed Rashid, Taliban: Militant Islam, Oil, and Fundamentalism in Central Asia (New Haven, Yale University Press 2000) 56-7, 149.

(31) Mahmud ibn Wali was assessing the Uzbek enemies for the ruler of Balkh, his patron. See: Edward Allworth, The Modern Uzbeks: From the Fourteenth Century to the Present: A Cultural History (Stanford University, Hoover Institution Press, 1990) 8-10.

(32) This is confined to books published in the 1990s and afterwards. Result gathered from Google Scholar citations feature, 7 October 2014. Coll has very little to say about Afghanistan’s ethnic Uzbeks, aside from some basic narrative of Dostum’s past. Rubin has a little more to say about Uzbeks, but it is mostly just neutral mentions of the role they played in the wars of the 1980s and early 1990s. See: Steve Coll, Ghost wars: the secret history of the CIA, Afghanistan, and bin Laden, from the Soviet invasion to September 10, 2001 (New York, Penguin Press 2004); Barnett R. Rubin, The Fragmentation of Afghanistan: State Formation and Collapse in the International System, 2nd edition (New haven, CT, Yale University Press 2002).

(33) For example, see: Anna Badkhen, ‘“In My Father’s House They Gathered All the Women into One Room”: Visiting the victims of Afghanistan’s revenge rapes,’ Foreign Policy (17 April 2010). Similarly, a BBC report chose to use the example of a Hazara attack on Pashtuns. See: ‘UN seeks to end Afghan abuses,’ BBC News (7 March 2002). In a report that did not draw from the Human Rights Watch report, Junbish soldiers (ie, Uzbeks) were singled out and when Tajiks were mentioned, they were portrayed as police officers unable to stop Junbish attacks. See: David Filipov, ‘Warlord’s men commit rape in revenge against Taliban’, Boston Globe (24 February 2002). Similarly, see The New York Times article that singled out Uzbeks: Dexter Filkins and Barry Bearak, ‘The Pashtuns: A Tribe Is Prey to Vengeance After Taliban’s Fall in North’, The New York Times (7 March 2002).

(34) An example here is a BBC article wherein it is stated that many people from all of Afghanistan’s ethnic groups have been accused of human rights violation, citing Human Rights Watch, but then singles out the Hazara Mohaqeq and the Uzbek Dostum from amongst these: Pam O’Toole, ‘Afghan poll’s ethnic battleground’, BBC News, 6 October 2004.

(35) Antonio Giustozzi, ‘Don’t call that warlord a warlord’, Foreign Policy, 25 February 2010.

(36) Source: A search using the terms shown in the graph, from 1990-2008. ‘Hazara warlord’ is not used enough to be tracked by the Ngram search. The y-axis on this chart is the percentage of the terms’ usage in all books. See here.

(37) For example, see: ‘Afghan presidential contenders sign unity deal’, BBC News, 21 September 2014.

(38) ‘Afghan presidential election audit completed’, BBC News, 5 September 2014.

(39) Antonio Giustozzi, ‘The Ethnicisation of an Afghan Faction: Junbesh-I-Milli from its Origins to the Presidential Elections’, Crisis States Programme Working Paper, No 67, Crisis States Research Center (September 2005) Available online here.

(40) Source: a search of Google Scholar up until 2014 using the terms “Uzbek,” “Tajik,” and “Pashtun” and “Hazara” combined with “warlord.” The results show a similar ratio to the Ngram search above. Hazaras a very rarely labeled as ‘warlords’ while Uzbeks are about four times more likely than their Pashtun and Tajik counterparts to be similarly labelled in the search. However, the civil war in Tajikistan has skewed the results: the use of “Tajik warlord” in the scholarly literature on Afghanistan is even lower than the search indicates.

(41) Source: a simple counting of all Google returns up until September 2014 using the above terms shows Uzbeks are three times more likely to be labelled as warlords when compared to Tajiks and Pashtuns.

(42) ‘Blood-Stained Hands: Past Atrocities in Kabul and Afghanistan’s Legacy of Impunity’, Human Rights Watch report, 7 July 2005; ‘Casting Shadows: War Crimes and Crimes against Humanity, 1978-2001’, The Afghanistan Justice Project, 2005; ‘[UN Mapping Report]’, unpublished UN report on human rights abuses in Afghanistan from 1978-2001, completed in late 2004, leaked online five years later.

(43) See, for example: Sari Kouvo, ‘After two years in legal limbo: A first glance at the approved ‘Amnesty law’’, Afghanistan Analysts Network, 22 February 2010; ‘Afghanistan: Repeal Amnesty Law’, Human Rights Watch, 10 March 2010.

(44) Source: Survey of Google returns up until September 2014 shows Uzbeks are three times more likely to be labelled as warlords when compared to Tajiks and Pashtuns.

(45) These websites were generally out of the mainstream media, and included such outlets as the World Socialist Website (here or here), the Revolutionary Association of the Women of Afghanistan (RAWA) (here or here), Counterpunch and the Yellow Times.

(46) Paul Watson, ‘Tide turns with aid of defections’, Los Angeles Times, 13 November 2001.

(47) This is revealed by a simple google search across the regular google platform using a combination of terms and searching within certain date ranges. One journalist even uses The Los Angeles Times’ full description of “one of Afghanistan’s most notorious turncoats” a decade later in her book without citation. See: Anna Badkhen, Peace Meals: Candy-Wrapped Kalashnikovs and Other War Stories (New York, Free Press 2010) 27.

(48) The same Google Search exercise can be repeated above by replacing ‘notorious’ with ‘treacherous/treachery’ or ‘brutal/brutality’, with the same results. This does not work for other commanders. The exception of here is the usually negatively portrayed Gulbuddin Hekmatyar, an active member of the insurgency. For a positively portrayed commander, the deceased Ahmed Shah Massud is a good example: ‘freedom fighter,’ ‘legendary,’ ‘hero’ and, especially, ‘charismatic’ work in the same pattern.

(49) Note: The source material criticises commanders of all ethnicities and affiliations, but the end result in the writings of journalists and analysts disproportionately focuses criticism on Rashid Dostum and on Uzbeks in general. In the last few years there have been few references to Uzbek warlords, mostly because so few Uzbek commanders have managed to maintain their position of power or to rise from obscurity. The only references in the last three years to Uzbek ‘warlords’ other than Dostum are to Qazi Kabir, Ahmad Khan Samangani, Piram Qul, Abdul Malik Pahlawan, and Abdul Rauf Ibrahimi. Sources: Tom A Peter, ‘Suicide attack in Afghanistan’s north signals broader reach of Taliban’, Christian Science Monitor, 21 February 2011; The Associated Press, ‘Suicide bomber kills top Afghan MP, 22 others at wedding’, CBC, 14 July 2012; ‘Ziayee, Al-Haj Peram Qul Zyaee Piram Qul’, Afghan Biographies, 21 October 2011; Fabrizio Foschini, ‘Ashura Attacks (3): A new type of violence in Afghanistan’, Afghanistan Analysts Network, 7 December 2011; Hamid Shalizi, ‘Afghan parliament elects ex-warlord as speaker’, 27 February 2011.

(50) Magsie Hamilton-Little, ‘Taking tea with Afghanistan’s most fearsome warlord, General Abdul Rashid Dostum’, The Telegraph, 29 July 2012.

(51) Autocomplete search

(52) 8 October 2014 on with no cache, no cookies and using a variety of internet anonymisers. This is in order to hide my personal browsing history, which Google does take into account when providing search returns.

(53) This was determined by conducting an internet search combining the names of these and other commanders with ‘warlord’ in combination with their ethnicity (first search) and without their ethnicity (second search). From this it was clear that Rashid Dostum was more than ten times more likely than Massud, Ismail Khan and Hekmatyar to be identified both as a warlord and by ethnicity, and five times more likely that Sayyaf. Source: Google Scholar index search, 9 October 2014.

(54) Discussions in person and online with a variety of foreigners, mostly analysts, but including a few journalists, 2007-2014.

(55) Craig Charney, Radhika Nanda and Nicole Yakatan, ‘Voter Education Planning Survey: Afghanistan 2004 National Elections’, The Asia Foundation (July 2004), p 85.

(56) For example, see these publications, the first by a journalist and the second by an academic: Ann Marlowe, ‘“Warlords” and “Leaders”: The hidden agendas behind press coverage of the Afghan war’, Personal Blog, re-posting of article from National Review Online, 18 February 2002; Brian Glyn Williams, The Last Warlord: The Life and Legend of Dostum, the Afghan Warrior who led US Special Forces to Topple the Taliban Regime (Chicago, Chicago Review Press 2013).

(57) See, for example, the complaints by Dostum supporters of anti-Uzbek racism in Afghanistan: Asla Aydintasbas, ‘Gen. Rashid Dostum: The Uzbek warlord, and Afghanistan’s new interim deputy defense minister, sounds enlightened, but can he walk it like he talks it?’, Salon, 9 January 2002.

(58) Human Rights Watch is notable for objectively describing human rights violations by all parties to the conflict in Afghanistan. See their reports and press releases here.

(59) On this incident, see: Robert Peszkowski, ‘Reforming Jombesh: An Afghan Party on its Winding Road to Internal Democracy’, Afghanistan Analysts Network, 31 August 2012.




Dostum ethnic groups image influence Pashtuns reputation Tajiks Uzbeks war crimes Warlords


Christian Bleuer

More from this author