One of the great difficulties with discussing the event known as the Zunghar Genocide of 1757 is that almost all of the English-language historiography on it points back to a single secondary source, that being Peter C. Perdue's China Marches West (2005). It is a fantastic book with great information and great analysis, but the Zunghar Genocide makes up only a very small portion of it, and to draw too many conclusions can be quite risky. I say this in order that it is made as clear as possible that I am also drawing on this particular source and will not be trying to draw too many conclusions beyond what Perdue, who has done far more primary source research on this topic than I am likely to, will.

The Zunghar Genocide was the final blow inflicted by the Qing in a series of conflicts, known collectively as the Zunghar-Qing Wars, that lasted nearly seven decades, beginning with the invasion of the Khalkha khanates by Galdan Khan in 1688. We do not have any good estimates of the extent of the genocide, and Perdue notes that his figures come from a source nearly a century after the events, the Shengwuji of Wei Yuan, written in the mid-nineteenth century. The Shengwuji, which was written based on official Qing reports, estimated that the Zunghar population stood at around 600,000, and that over the course of 1757, around 40% died of smallpox (accelerated by the advance of Qing troops and the displacement of Zunghars from their lands), 30% were killed directly by Qing soldiers, and 20% sought refuge with the Russians and the Kazakhs, the remainder being enslaved – though there are reasons to doubt some of these proportions, which I will get on to. The Zunghars as a coherent people ceased to exist, with the survivors being subsumed into the broader Mongol and Kazakh groups.

How did it get to this point? The Zunghar Khanate was arguably the last great nomadic steppe empire, emerging out of the ruins of the Oirat confederation in the 1630s, in parallel with the unfolding collapse of the Ming. While their main territorial core was somewhat limited, consisting principally of Zungharia (the northern, grassier portion of what is now Xinjiang) and western Mongolia, their strategic position between the Qing Empire to the east, Tibet to the south, the Tarim Basin to the southwest, the Kazakhs to the west and the Russians to the north made them a key player in the geopolitics of Early Modern Asia. Batur, the leader of the Choros tribe, had attempted to re-establish some kind of unified Oirat state after the dissolution of the Four Oirat in the 1630s, but died in 1653, cementing the fragmentation of the Oirat tribes between a series of sometimes-cooperating, sometimes-rivalled entities.

Before we continue, it is worth going into detail on the nature of identities on the steppe and of political structures. The key thing to note here is that when I refer to 'Kazakhs' or 'Oirats' or even 'Mongols', I am referring to political, not ethnic identities. The notion of ethnicity among steppe peoples is something that came about as a product of prolonged subjugation by sedentary powers, such as the Qing (and latterly the Republic of China and People's Republic) and Imperial and Soviet Russia. It was theoretically possible, for instance, for a particular identity to dissolve simply due to the deaths of prominent political leaders and consequent collapse of cohesion among the group. The other important aspect is the nature of political leadership in the wake of the fracturing of the Mongol Empire. While few groups on the steppe considered themselves 'Mongols', virtually all of them subscribed to the idea that the only people entitled to undisputed authority were Chinggisids, patrilineal descendants of Chinggis Khan. Relations by marriage were a source of prestige, but not necessarily of entitlement to rule – though this did not stop Timur in the fourteenth century, who married the daughter of a Chinggisid prince and whose descendants, self-styled Gurkani ('sons-in-law [of Chinggis]'), came to rule northern India as the Mughal emperors. But this matter did get in the way of chiefs such as Batur who sought to consolidate their control, as it created a limit to their official legitimacy that could and would remain in place for any successor.

This digression is relevant for a few reasons: most importantly, it will help contextualise the implications of the genocide when I get to it later. But also, it shows how the Zunghars as a people came into being. While the exact point at which Choros, Dörbets and so on morphed into Zunghars is somewhat hazy, the fact that it did happen was the product of a relatively fungible landscape of identities. It also illustrates why the Zunghars were so dangerous to the Qing, because what Galdan Khan did – or rather what was done for Galdan – had the potential to radically reshape the politics of the steppe.

Galdan (born in 1644) had been training as a monk under the Fifth Dalai Lama at Lhasa since the age of seven, and continued living in monasticism, leaving the succession squabbles in the wake of the death of his father, Batur, up to his brothers. Inconveniently, however, in 1670 the succession squabbles had led to Batur's designated heir, Sengge, being assassinated, and Galdan returned to Zungharia to claim power for himself, primarily in competition with his father-in-law, Ochirtu, whom he defeated and killed in 1678. He did not lose his monastic connections, however, and the Dalai Lama gave him the title of 'Boshoktu Khan', allowing him to claim sole leadership of steppe tribes without Chinggisid bloodline. It also gave form to a religious aspect to Galdan's ambitions, as he saw his objective as not simply the political unification of the steppe, but also – or even instead – the consolidation of the religious authority of the Dalai Lama over the khutukhtus who served as the regional and local religious leaders for most of the eastern Mongols. Seizing on disturbances in the Khalkha khanates as a pretext, he invaded eastern Mongolia in 1688, driving many Khalkhas into outer Manchuria, then hotly contested between the Qing and the Russians, and temporarily brought the eastern steppe under Zunghar control.

However, this brought Galdan and the Zunghars prominently onto the Kangxi Emperor (r. 1661-1722)'s radar, and after concluding peace with the Russians in 1689, the Qing monarch turned his attentions to Mongolia, securing the fealty of the defeated Khalkha khans. A Qing army intercepted a Zunghar army heading for northern China at the Battle of Ulan Butong in 1690, inflicting heavy losses, and in 1696, a small Zunghar army consisting of most of Galdan's core retinue was virtually annihilated by a Qing expeditionary column, and he died suddenly, possibly of smallpox, in early 1697. His rival, Tsewang Rabdan (r. 1697-1727), took control of the khanate, but taking the title of Khong Tayiji ('Prince' or 'Regent') as he lacked either Chinggisid bloodline or Gelug recognition – the overtly religious element of the Qing-Zunghar conflict would be more subdued from here on out. This marked the beginning of a 'cold war' of sorts between the Qing and the Zunghars, with key moments including a Zunghar invasion of Tibet in 1718 and Qing counter-invasion in 1720, the destruction of a Qing army under Furdan at Zunghar hands in 1731, and a Zunghar civil war following the death of Galdan Tseren, Tsewang Rabdan's successor, in 1745. Throughout this period, preparations for a grand invasion of Zungharia by the Qing were slowly developed by the Yongzheng Emperor (r. 1722-1735), and under his son the Qianlong Emperor (r. 1735-96/99), would be employed in full, initially to support a rival claimant to Zunghar leadership, Amursana, against the de facto prince, Dawaci. However, almost as soon as he was placed in charge, Amursana revolted from Qing overlordship in 1756 along with the Khalkha prince Chingünjäv.

The revolts of Amursana and Chingünjäv would mark a decisive turning point in Qing policy towards the Mongols and Oirats. In Mongolia, after the capture and execution of Chingünjäv, authority for the selection of khutukhtus was transferred to the central college of lamas in Tibet in order to prevent nepotistic appointments (as Chingünjäv's rebellion had been granted legitimacy by the Jebzongdanba Khutukhtu, the principal regional Buddhist cleric and who just so happened to be Chingünjäv's brother); while in Zungharia, the Qing launched an expedition with some 50,000 troops, quickly routed Amursana's army, and embarked on the campaign of genocide that is now under discussion here.

No single factor accounts for why the Qianlong Emperor considered genocide the path to take. Earlier and later campaigns against the Jinchuan, a group that might be considered similarly 'nonconformist' (being Tibetan-speakers who did not recognise the authority of the Dalai Lama), lacked such an overtly genocidal aspect, and while civilian massacre and similar war atrocities had characterised parts of the Qing conquest of China in the 1640s through 1660s, these atrocities were again not genocidal in intent. Ideologically, the Zunghars represented a challenge to Qing rule that threatened their coalescing conception of universal emperorship. In order to be entitled to rule all the Mongols, no alternative centre of political power among the Mongols could exist, and the Zunghars represented such a rival centre. Pragmatically, the intractable resistance of Zunghar leaders and the constant failure – or at least, lack of longevity – of attempts at achieving peaceful settlement must have caused sufficient frustration to the three successive Qing emperors who oversaw the conflict that such a policy of massacre became conscionable. This is not to excuse the actions of the Qing emperors or their armies, of course, but rather to explain a course of action that, for the Qing emperors at least, was decidedly uncharacteristic.