Eastward across the Western Sea: The Indian Oceanic Trafficking of Africans into China

Abstract

Despite its greater extensiveness in comparison to the ferrying of their West African counterparts across the Atlantic Ocean into bondage in the New World, the history of the extraction of East Africans to serve as slaves in the various lands that ring the Indian Ocean is barely known to most of us. Particularly as Westerners, our knowledge of even the sketchiest outlines of the latter phenomenon pales before what we know intimately about the intricacies of the former. Furthermore, unknown altogether to too many is the fact that—from at least as early as the eighth century of the Common Era—these East African slaves were exported as distantly as China. Based principally on the pertinent Chinese sources, this study raises and investigates three fundamental questions concerning this conveyance of East Africans into China. First, who were the original enslavers of these East Africans and thus the prime purveyors of their entrance over the centuries into bondage in China? Second, how—that is, by land or by sea—and under what auspices were these East Africans typically transported from their homelands mainly along the eastern African coastline to a locale as far away as China? Third and finally, what was the probable fate of these East Africans once they had arrived in China? The essay concludes with a consideration of the consequentiality of this largely overlooked and therefore unheralded saga amidst the history of Chinese institutionalised servitude overall.

A key feature of modern historical inquiry is that we must continually revise our interpretations of the global past. One result of this enterprise has surely been the repeated discovery that contact between widely dispersed and seemingly unrelated peoples almost always seems to have occurred earlier than previously thought. Nonetheless, when addressing the subject of the enslavement of Africans by non-Africans, we still somehow mentally resist situating slaves originating from Africa in a context as physically and culturally remote as China.

Rather, an overriding preference for geographical proximity compels us to regard any place much nearer to Africa than antipodean China as a far more viable point for contact and thus a more credible candidate for the dubious recognition as a bastion for the trading in African slaves. In addition to India, more likely “Asian” prospects include the polities that ringed the Persian Gulf as well as the more populated zones of the Arabian Peninsula. Our hunch is confirmed by the fact that African slaves were at all times more numerous in the western portion of the Indian Ocean littoral zone than in its eastern portion.¹ Over a period extending back some four millennia or more, the slave-trading inhabitants of these nearby territories took advantage of their proximity to a seemingly endless supply of Africans who could be lured or captured and pressed into bondage.² As a result, transoceanic slave trafficking arose very early on the eastern side of the African continent, where Arab and Persian merchants sought slaves from the eighth century CE, some seven centuries before the advent of the transatlantic slave trade from West Africa.³

The distance between China and the source of African slaves is not all that militates against Chinese involvement in African slavery. The unfamiliar terminology used in classical Chinese texts to describe black Africans also clouds the question of the presence of enslaved Africans in China. The earliest designation that Chinese sources apply to Africans is Kunlun. ⁴ Kunlun as a descriptor for human beings is derived originally from a place-name, even as it inconveniently applies to a great number of places.⁵ Still more problematically, Kunlun was the same name that the Chinese had already for centuries arbitrarily affixed to a profusion of non-African peoples, comprising Malays, Negritos, Papuans, Melanesians, Khmers, Champans, Srivijayans, Javanese, Borneans, Ceylonese, and Indians of South Asia, as well as a host of West Asian (or Middle Eastern) peoples. The only qualification for Kunlun status was relative darkness or blackness of skin—at least in comparison to the observer writing the classical Chinese.⁶ This extremely low threshold qualified even the occasional Chinese as a Kunlun. ⁷

Consequently, ascertaining the geo-ethnicity of Kunlun referred to in the extant Chinese sources is challenging. Foremost among the difficulties is determining when a member of this designated catchall group subjected to enslavement in China was actually an African as opposed to being a dark- or black-skinned member of one of these other groups. Indeed, based on the documentation available to us, ascertaining when the first Africans arrived as slaves in China in itself will always be a somewhat inexact science. On the one hand, from about the tenth or eleventh century CE onward, Chinese references to Kunlun may more often than not be persuasively interpreted as equivalent to references to Africans. Yet on the other hand, as we shall see, complicating the endeavour are the many distinct Chinese names other than Kunlun—with all of them to a greater or lesser degree disparaging—that had also emerged as designations for dark- to black-skinned individuals by this time.⁸

Despite such pitfalls, this article argues that from fairly early on in Chinese imperial history, the term kunlun—almost invariantly in combination with the suffix nu, meaning “slave”—emerged as the principal descriptor for enslaved individuals of African origin or descent. As such, it later became not only the chief designation for the largely African captives brought in bondage by the Portuguese and other Europeans during the Ming (1368–1644) and Qing (1644–1911) dynasties, but also the normative template for their Chinese reception. However, the objective here is neither to supply a localised catalogue of such names in the early modern era nor offer a disquisition on race awareness in modern China in the manner of either Peng Hui or Frank Dikötter.⁹ Instead, this article's first aim is to demonstrate that Kunlun was without question the main or core name used for the plausible as well as real African, regardless of period. This continuity in terminology underlines that even as European regimes of bondage expanded into China, they were interpreted within long-standing Chinese frameworks. The second aim is to reconstruct (as far as the surviving sources permit) the work that Kunlun nu were assigned and their level of agency within Macanese and adjacent Cantonese society. Such questions have yet to be raised in the largely positivistic Sinological scholarship on the topic. Before delving into such sociological questions, however, we need to establish some basic facts about enslaved Africans in China and Chinese knowledge of Africa and Africans, topics that to this day still do not garner the attention they deserve from historians of premodern Afro-Eurasia.

Siting and Citing the African Presence

As a reference to the black-bodied, if not always to the enslaved, Kunlun appears in Chinese sources datable to as early as the first centuries of the Common Era. This earlier tradition is essential for understanding attitudes in the early modern period, as these classical texts were widely read by later scholars working within the Sinographic tradition. Some modern authors suggest that some of these very first historical Kunlun were Africans.¹⁰ This argument hinges on a claim that is difficult either to prove or disprove—namely, that mariners from Indonesian kingdoms such as Srivijaya and Java actually preceded the Arabs, if not the Persians, in their participation in the East African slave trade.¹¹ This theory is additionally premised on the far less contentious idea that captives of African origin might well have entered China as slaves before the Chinese themselves had acquired a firm knowledge of the existence of Africa as a place. In the absence of either archaeological or written evidence, this is entirely possible. However, it remains for now unresolvable beyond all doubt. Therefore, with our interests better served by pursuing what we can know, by way of background let us briefly consider the question of when it was that the Chinese first discovered Africa, and how they described it in texts that would go on to cast a long shadow into the early modern period.

Based on the Chinese penchant for committing encounters with the exotic or unusual to writing, we are on safe ground in expecting the discovery of any African site to have been recorded. In this connection, albeit tenuous, some latter-day scholars have claimed to have found the first reference to Africa in a Chinese text in the Weilue (Brief history of Wei),¹² a work of the third century CE attributed to a mysterious author named Yu Huan, about whom nothing further is known.¹³ Therein, “Africa” is arguably first referred to in the form of the isolated appearance of the name Wuchisan, which just might be the city of Alexandria in Egypt.¹⁴ Still, whether Wuchisan is truly identifiable with Alexandria remains debatable, and thus we find that we can neither reliably nor conclusively surmise that the Chinese had garnered any solid awareness of the existence of Africa before the ninth century CE.

Our much sought-after, first-ever reference in a Chinese source to an inarguably African territory only appears six centuries later in a work by the Tang-dynasty author Duan Chengshi (ca. 803–63 CE).¹⁵ In his Youyang zazu (Miscellany of tidbits from Youyang mountain cave) written around 855 CE, Duan offers a singular description of Bobali, which scholars now concur to have been Berbera, or what is today the greater part of the northern coastline of the peninsular horn of modern Somalia.¹⁶ In customary fashion, Duan Chengshi's description of Berbera begins with where he believes it to have been located, before discussing its population.¹⁷ Here, however, he makes no mention of what then as well as now would have been the most fundamental distinguishing trait of the Berberans—that is, their darkness, indeed blackness, of skin colour.

Following Duan Chengshi's account, there is no explicit reference in a Chinese text either to Africa or to probable Africans for more than a century, when the mandate of rule over China had passed on to the succeeding Song dynasty (960–1279 CE). The reference itself is made only in passing and with such unremarkable soberness as to seem almost incidental, and yet we find that it has the unassailable distinction of appearing in an official source—the Songshi (History of the Song). Of a Persian tribute delegation that arrived at the imperial court of the capital Kaifeng in order to commemorate the accession of the second ruler of the dynasty in the previous year, we are informed that:

In 977, Envoy Pusina, Vice Commissioner Mohemo (Mahmud? Muhammad?), and Administrative Assistant Puluo and others made an offering of tribute [to the court] of the goods of their locality. Their servants, having deep-set eyes and black bodies, were called Kunlun slaves. By imperial decree, in return, these envoys were given suits of garments with lining, utensils, and currency; their servants were given variegated silk fabric, with differences [in each case].¹⁸

Were the Kunlun described here Africans? Although terse, the passage is unquestionably interlaced with the kinds of hints that encourage us to think so. For instance, their designation as “servants” of the Persian emissaries suggests that they had long been part of the retinue, rather than having been acquired during their voyage to China. There is also the clear implication that, like the ambassador, they would be returning to Persia. Thus, whereas these particular slaves cannot be deemed categorically to have been Africans, the extraction of slaves from Africa by Persian merchants that had been ongoing for several centuries encourages the attribution of an African identity to the Kunlun nu above described.¹⁹

Another Songshi entry not only reaffirms Chinese knowledge of Africa as a geographical entity, but also provides us with a vivid description of their exposure to persons who were unmistakably Africans, even if they were not slaves per se. Through this latter entry, we actually learn of Africans travelling to China on two related occasions but receive no inkling that Chinese were yet trekking to Africa. At another juncture in the Songshi, we find a description of a country or perhaps a city called Cengtan, which appears to be a variant of Cengba, an eleventh-century term for Zanzibar—a place-name that in Arabic originally meant the “Region of the Blacks,” that is to say the area from south of the Juba River to Cape Delgado along the eastern coast of Africa.²⁰ The Songshi further explains:

Cengtan is in the Southern Sea, with its town about twenty li [or seven miles] inland from the coast. In 1071, Cengtan brought gifts to our court for the first time. Traveling by sea, with the favorable [monsoon] winds, the envoy took 160 days. Sailing by way of Wuxun [in the vicinity of today's Muscat, Oman], Kollam [in India], and Palembang [in Indonesia], he arrived at Guangzhou.²¹

Beyond its confirmation of a preference for travel to China by ship, which either may or may not mirror the previous instance of the Persian mission of 977, in the above description we learn additional details about Cengtan. For instance, we are informed that its “language sounds like that of the Arabs” and that its “climate is warm year-round.”²² Most important among these details, however, is that the initially nameless envoy of Cengtan becomes named: “In 1083, the envoy Protector Commandant Cengjiani came [to China] again [bearing gifts] to court. [Our emperor] Shenzong [r. 1067–85 CE], recognising the extreme distance he had travelled in returning, beyond presenting him with the same gifts as before, added 2,000 taels of silver.”²³

We cannot know if the distinguished Cengjiani was black, or even if Cengtan in particular or Africa in general was his birthplace. Yet, even if he lacked African ancestry or provenance, we may safely assume Cengjiani to have come as directly as he could to China from Africa, given the difficulty and importance of the mission. Moreover, from the exalted title the Song emperor bestowed on him, the latter part of which—“commandant” (langjiang)—was granted only occasionally to aboriginal chieftains of the southern and western borderlands, we can detect that the Chinese regarded this emissary from Africa with high esteem.²⁴ However, most crucial for our purposes is that his name itself in Chinese (ceng-jia-ni) appears to be a phonetic approximation of the Arabicized Persian term zanj (also zandj or zang or zeng) or possibly the Latinized Arabic zinj, which was likely meant also to denote his Zanzibarian place of origin.²⁵ Of course, given its literal meaning in Arabic as “the blacks,” we find that this same phoneticised corruption of the very term used by Arab traders for those they enslaved would eventually enter into and become a fixture of Chinese parlance as a designation for the increasing numbers of Africans they themselves would incrementally encounter.²⁶

This is not the only example of people from Zanj who found themselves in China. Moreover, not all this trafficking in the Zanj was in Persian and Arab hands. For instance, sources describe the occasional delivery of Zanj captives as curios to the imperial court, beginning in the eighth and ninth centuries. These sporadic deliveries of Zanj slaves were usually made by tribute missions dispatched to China from near-neighbouring countries in the South China Sea. An official entry in the Xin Tangshu (New history of the Tang) on the reception of the Javanese tribute delegation of 813 CE serves as a case in point. Here, we are informed that the emperor Xianzong (r. 806–20 CE) “received four Zanj (sengqi) slaves, a five-coloured parrot, human-faced birds, and the like.”²⁷ People from Zanj, if the identification is indeed correct, were clearly not all that uncommon in the region.

In the three centuries following Duan Chengshi's initial and highly impressionistic description of Berbera, Chinese knowledge of Africa—either derived directly from dealings with first the Persians and subsequently with the Arabs themselves or channelled through mostly Chinese mariner go-betweens—increased exponentially. Owing to instability on the borders of the empire during the later Song, the customary Central Asian Silk Road land routes became impassable. As a result, Chinese knowledge of nearby and distant maritime states and their sundry peoples increased significantly.²⁸ For example, the Lingwai daida (Information on the lands south of the Wuling mountains) by Zhou Qufei (jinshi 1163 CE) and the Zhufan zhi (Description of foreign peoples, or records of barbarous nations) by Zhao Rugua (1170–1231 CE), completed in 1178 and about 1225 CE, respectively—include numerous geographical treatises of varying lengths and levels of detail, which describe the different topographies, products, and peoples of East Africa.²⁹ We should by no means overlook the biases that permeate and misinformation that lingers in these two books and the less comprehensive works that survive. Yet, despite their shortcomings, these revealing opuses of the twelfth and thirteenth centuries on the world beyond China greatly surpass in accuracy the mostly outlandish, ill-informed, and fantasy-riddled writings that preceded them.³⁰

With the defeat and supplanting of the Song by the succeeding Yuan dynasty (1279–1368 CE), China was incorporated into the Mongol empire. The resulting Yuan regime maintained an extensive network of foreign relations and, in the spirit of the Mongol drive for conquest and profit, encouraged the geographical study of the world, including the so-called Western Sea (that is, the Indian Ocean).³¹ Indeed, recent research has largely debunked the landlubber image of the Mongols as vaunted horsemen but haplessly inept sailors.³² Furthermore, the Mongol conquests reopened land routes that had declined during the Song, allowing Zanj, whether free or enslaved, to enter China overland.³³ The unprecedentedly multicultural environment fostered especially under the emperor Khubilai Khan (1215–94; r. 1260–94 CE) also piqued an interest in all things foreign.³⁴ Indeed, the overall trend towards greater geographical knowledge in the Song period aligned especially well with Mongol self-interests, as can be seen in Yuan-era cartography.³⁵ One of the best surviving examples of this tendency exists in the form of an off-centre depiction of the southernmost tip of Africa by the notable geographer Zhu Siben (1273–1337 CE)³⁶ drawn between the years 1311 and 1320 (figure 1).³⁷ Considering that the map was produced by an individual who had never visited the site he drew and who had only second-hand information to arrive at a representation, the rendering is stunningly accurate. Furthermore, upon closer inspection, an offshore island—just to the east of but still contiguous with the main shoreline of the continent—that appears to be Zanzibar jumps out.³⁸ Amidst the profusion of inconspicuous labelling, this island is strikingly and discernibly demarcated not as a place but as a people, sangce nu. This is presumably an approximation of zanj or “black”; nu, best interpreted in the plural in this context, can have no meaning other than “slaves.”³⁹

Figure 1. Map of Africa by Zhu Siben 朱思本 (1273-1337), in Luo Hongxian 羅洪先 (1504–64), Expanded Edition of Imperial Maps (Guangyu tu 廣輿圖), 1555. Image in the Public Domain.

This growing cartographical accuracy was further bolstered during the Ming thanks to the famed series of maritime voyages in the Indian Ocean by the imperial eunuch admiral Zheng He (1371–1435).⁴⁰ This involved hundreds of the largest known ships of the age crewed by tens of thousands of men who travelled as far as Hormuz, bringing back with them knowledge of these regions, their products and peoples that would continue to be influential throughout the Ming and Qing periods.⁴¹ During one voyage, they even escorted an embassy from Malindi (in modern Kenya) back to China.⁴² Thus, these seagoing expeditions of the early fifteenth century—before they were abruptly terminated under the sea ban (haijin), form the final element of the background to Ming and Qing encounters with Kunlun nu brought to China by the expansion of the Iberian slave trade and its associated regime of bondage.⁴³

Aboard Portuguese Carracks

With the cessation of the remarkable set of expeditionary voyages undertaken by Zheng He and enforcement of the sea ban, the trafficking of Africans into China—whether under Chinese or any other auspices—also ceased or, at the very least, slowed to a trickle.⁴⁴ However, the Chinese were soon to be exposed to Africans in quantifiable numbers in the tow of entirely different masters than in the past. These new masters of the Zanj were the imperialist-minded Portuguese and later Dutch and English who at the turn of the sixteenth century succeeded for a short stint at least in displacing the seafaring Arabs, Indians, and possibly Malays as keepers of and dealers in enslaved Africans throughout much of the expanse of the Indian Ocean.⁴⁵

By the close of the fifteenth century, the Portuguese were already leading participants in the slave trade along the African coast and they subsequently launched their forays into and throughout the Indian Ocean via their newly emergent outposts along the Malabar Coast (the southwestern coastline) of subcontinental India, which had been reached by the explorer Vasco da Gama (ca. 1460s–1524) in 1498 CE. Thereafter, the primary gateway for the new Portuguese presence and, concomitantly, the revived African one throughout the Western Sea was the city of Goa, which was famous for its slave market (figure 2).⁴⁶ From there, trade routes extended towards Ming China.⁴⁷

Figure 2. Anonymous. Black devil (Heigui 黑鬼). From Cai Ruxian 蔡汝賢, Portraits of Eastern Barbarians (Dong Yi tuxiang 東夷圖像), 1586. Image in the Public Domain.

Sailing from their subjugated Malayan entrepôt of Malacca, Portuguese ships first reached the great trading port of Guangzhou in the first decades of the sixteenth century, where they were greeted with suspicion bordering on outright hostility. Reports emanating from nearby Malacca describing the unfortunate and unseemly nature of Portuguese rule there tended to heighten Chinese wariness and hinder progress.⁴⁸ Locally and more directly, the Chinese found the ill-timed construction of a fort just downstream from Guangzhou by Simão Pires de Andrade (d. 1550s?) to be gallingly irksome.⁴⁹ Nor was the situation helped by the fact that Simão Pires de Andrade happened to be one of the first and most brazen Portuguese enslavers of Chinese.⁵⁰ These factors among others led Ming authorities to suspend official relations with the Portuguese in 1521, dismissing their ambassadors from the imperial court at Beijing and expelling their fleet from Guangzhou.⁵¹

In 1535, owing to the usefulness of the gunnery on board their caravels in eradicating the South China Sea piracy threat, Portuguese traders were finally and grudgingly granted permission by the Ming state to trade at, but not settle in Macau. Despite the prohibition on settlement, by 1640 Macau had a Portuguese population of some two thousand and, no less significantly, an enslaved population of about five thousand who were substantially if not all Africans imported from various parts of Africa.⁵² These enslaved Africans who accompanied their Portuguese captors to China represented the lowest rung of the social ladder in Macau as they did in other parts of the Iberian world, because, as the towering historian of Iberian Asia, Charles R. Boxer (1904–2000), put it: “The oft-made claim that the Portuguese had no colour-bar cannot be substantiated.”⁵³ Whereas the Portuguese kings generally maintained that religion and not colour should serve as the qualifier for full citizenship, this policy was widely ignored. It was particularly ineffective in the case of native populations of the Americas and those from sub-Saharan Africa. However, whereas there were legal prohibitions against the enslavement of the former, no such constraints applied to the bondage of the latter. Indeed, as Boxer astutely concluded: “The sugar plantations of Brazil, the household labour of the Portuguese settlers in three continents, and even to some extent the defence of their settlements, depended mainly on the strong right arms of their (principally African) slaves.” Although on a somewhat different scale, much the same could be said for Asia.⁵⁴

For their part, in their dealings with their only partly welcome foreign guests the members of Ming Chinese officialdom were wary of both European masters and African slaves, and reported persistently on the activities of both. In this context, a new term—actually, very likely a quite old epithet in the vernacular—began to appear as an alternative to Kunlun or Zanj. This was heigui, meaning quite literally “black ghost” or “black demon,” with occasional variations like heinu, “black slave,” and heiren, “black person.”⁵⁵ Heigui is distinctive not merely because it is so late in emerging in written texts, but because it is the first of the names that has no basis in geographical place (figure 3). We should also not doubt that it was highly pejorative. Therefore, ironically, even with enhanced amassed knowledge about Africa as a place in hand and more at their disposal than ever previously, the Chinese of late Ming times evinced a profound ignorance bordering on repugnance towards these black slaves.

Figure 3. Chen Xu 陳訏 (1650–1722). Chinese Men of the Tang Buying Foreign Slaves (Tangren mai fannu tu 唐人買蕃奴圖). Dated 1689. Reproduced in Chang Hsing-lang, “The Importation of Negro Slaves into China under the T'ang Dynasty (AD 618–907),” Bulletin of the Catholic University of Peking 7 (1930), 37–59. Courtesy of the British Library.

Yet, despite the rise to prominence of vernacular terms of derision like heigui, we find that an all-too-familiar term for the certifiably African blacks in the service of the Portuguese and other foreigners continued to hold sway with the Chinese literati. Perhaps nowhere is this retention better preserved than in the words of the geographer Wang Shixing (1546–98 CE), who—writing in his Guang zhiyi (Record of extensive travels continued)—observed while at Macau:

Moreover, there is a category of being aboard the foreign ships that is named “Kunlun slaves,” [the same] that the commoners call “black ghosts,” and whose entire bodies are as if coated in lacquer, which ends only with the whiteness of their two eyes. These beings heed only the master who feeds and clothes them, disregarding even the master's kin and friends. Whether they live or die is something that only their master can decree, and if the master orders that they should cut their own throats, then so they will do, with no thought to whether slitting their throats is deserved or not.⁵⁶

Much can be extracted from Wang Shixing's vivid description. Most salient, aside from the new attributions of unreflective obedience and wanton hyper-aggression, is just how imperceptibly little the Chinese reception, or perhaps more correctly apperception of these African slaves in their midst had changed, despite the passage of more than six centuries since the Song dynasty treated above.

Similar accounts of the Kunlun are to be found during the Qing dynasty (1644–1911). One of the most revealing of these is by the traveller and Guangzhou native Qu Dajun (1630–96), whose book Guangdong xinyu (News from Guangdong), dated 1680, records his first-hand experience of visiting Macau. At one juncture, he notes:

Of those that the Yue [i.e., Cantonese] merchants purchase to send along to the city of Guangzhou, all are yellowish-black [in complexion] with deep-set eyes. From the familiarity that comes with long stretches of contact, [these blacks] also become capable of Yue speech.⁵⁷

Interestingly, Qu states that not only did black slaves find their way to the provincial capital, but they were also sold there by both the Portuguese (from whom they had presumably been originally purchased from merchants or local potentates in the Indian Ocean), and local Chinese merchants. He also records other important facts, such as the advent of the Dutch, who evidently also traded—callously, from the author's perspective—in white slaves, not women, but young children: “Moreover, when the ships of the Red Hairs arrive, they frequently make gifts to others of small white children.”⁵⁸ Thereafter, however, Qu veers sharply into the realm of the fantastical, thereby seriously complicating, if not wholly undermining, our interpretation of his initial assertion, as he writes:

In height, they are limited to being perhaps a foot tall. Their faces as well as their hands and feet are like jade snow. Only their hair is auburn, and nothing more.

When they encounter people, they always remove their headdresses and kowtow. Their speech is extremely delicate; from the chirpings of birds, it cannot be distinguished. It is said that they are the offspring of the country of Small People, with males and females all being so. They are unable either to plough or sow. It is the case [instead] that they gain sustenance by harvesting and grinding into a powder a kind of white bean to eat that grows naturally in the wild.⁵⁹

On the one hand, by concluding with a harkening back to a trope of the existence of a remote land and fanciful people that is datable to the Western Han dynasty (202 BCE–9 CE),⁶⁰ Qu Dajun's record affords us a highly unexpected and eccentric but hardly unprecedented window onto the spectrum of early modern Chinese colour consciousness. Conversely, the question of whether we should hold his views to be representative of those of other, similarly exposed contemporary observers must remain open. This said, there is still a much larger point of discernment to be gleaned here: Qu Dajun makes a categorical distinction between the blacks held in bondage by Guangzhou merchants and the tiny white child slaves allegedly ferried into China by the Dutch. Whereas he is at best indifferent about the plight of the black slaves, Qu's portrait of the diminutive white child slaves, whether real or only imaginary, is flush with enchantment. Yet, regardless, we need little question which of these two broad varieties of slaves had collectively come to represent the truer face of the enslavement of foreigners in Qing China by Qu Dajun's time. Especially upon the entrance of the Europeans themselves, surely the stolid and muscular blacks rather than any questionable varieties of miniature whites were more prevalently imported as slaves, despite Qu's fixation on and seeming predilection for a particularly entrancing breed of the latter, if it had in fact ever existed.

If nothing else, then, we may conclude from the scattered accounts written during the Ming and early Qing that despite the long stretch of time between that period and the first Chinese encounters with people from Africa there was considerable continuity in terms of terminology and attitudes. Chinese colour consciousness is clearly not a modern phenomenon. There were subtle changes, however, with the advent of the Portuguese in terms of the frequency of such contacts and their appearance in the written record, as well as a greater explicitness about the relationship between skin colour and bondage.

Unanswered Questions of Labour and Agency

Despite the foregoing conclusions, many questions remain regarding the untold numbers of Africans who, preponderantly as slaves, were taken from the eastern shores of their native continent to China during the millennium or so before the advent of the transatlantic slave trade and for several centuries afterwards. In particular, it remains hard to ascertain the collective function of those who were subsumed along with others under the variegated rubric of Kunlun and other catchall terms. Especially within a state that at no stage in its history relied on a slave class or its labour as the dominant mode of production (if the standard Marxian economic histories are to be believed) and had a very large population by premodern standards providing an almost infinite supply of cheap labour, it is far from clear what the motive behind the presence of foreign slaves in general and African slaves in particular might have been.⁶¹ Here, some premodern comparisons may shed some preliminary light, especially when read alongside the accounts of Wang Shixing and Qu Dajun.

For instance, we can observe that neither in the lands under Islamic dominion nor in China were slaves at any point the main basis of production, as they had been, for example, in the Roman republic and empire. Referring specifically to the epoch when the caliphate of the Abbasids was at its greatest territorial extent in about the year 850, the historian Bernard Lewis observed that whereas they were mainly utilised domestically and militarily (especially, in the latter instance, in the formation of the Mamluk armies), slaves were occasionally put to work in mining, fleet production, and dredging.⁶² The unexceptional nature of such “societies with slaves” as well as their ability to exist close to more recognisable slave societies is further underlined by recent studies of slavery in urban contexts in the Americas like Mexico City, which are likewise hard to integrate in the traditional Marxian framework.⁶³ As we have seen, albeit largely under personal auspices as guardsman and retainers, the heigui or “black ghosts” were to be found living in close proximity to their Portuguese masters in urban contexts in Ming China. The same was likely true under Chinese masters in Guangdong. Furthermore, Chinese sources of Song, Yuan, Ming, and Qing dynastic times indicate that those Africans entering China who were misfortunate enough not to be gifted as tribute were subjected to a fate very similar to their brethren held in Arab, Mongol, or Portuguese captivity. For instance, we learn that Kunlun—perhaps already becoming incrementally more African than Malay—toiled for their Chinese masters as stevedores, divers, and menders of the hulls of damaged or grounded ships.⁶⁴

There are also continual references to their skill as bodyguards, mariners, and swimmers. For instance, in the aforementioned Guang zhiyi we learn that:

The Kunlun by nature brandish blades and are skilled at killing. If the master departs and orders his slave to guard the gate, then whether fire or flood occurs, he will perish before escaping. If someone slightly nudges the door-bar or the bolt to a box wherein valuables are kept, with no deliberation on whether he is a thief or not, the slave then simply kills him. These slaves are also skilled at diving and, by using cords tied around their waists, they enter the waters to fetch things. These slaves are purchased by the head, at a cost of fifty or sixty pieces of gold each.⁶⁵

Such reports no doubt found their way to Chinese readers, some of whom may have sought out Kunlun nu for their own households or ships.

Of course, the critical question of the precise function of African slaves within traditional Chinese society is complicated all the more by the fact that China had from prehistoric until fairly recent times practised and condoned endogenous forms of bondage (that is to say Chinese did for generations enslave other Chinese), and it is likely that customs surrounding this convention also shaped the lives of Kunlun nu. Furthermore, given the plentiful supply of local labour, including in an extreme form of bondage, it remains unclear why Chinese would resort to foreign slaves at all. Beyond the reality or perception of their aforementioned martial and maritime skills, a motivation may have been their role as exotica within a wealthy household serving to satiate an overpowering fetish for possessing the unusual. Such things certainly happened at the courts of premodern Europe.⁶⁶ To acknowledge this possibility, on the one hand, risks trivialising the tragically unrecoverable experiences of these long-departed slaves. However, on the other hand, not to acknowledge it results only in the wilfully irresponsible avoidance of the potential truth. In sum, from the earliest periods of Chinese imperial history, the term kunlun—almost invariantly in combination with the suffix nu, meaning “slave”—emerged as the principal descriptor for enslaved individuals of African origin or descent, as well as smaller numbers of dark-skinned people of different origins who were occasionally lumped together with Africans. During the Ming and Qing dynasties, this would later be supplemented by popular terminology, such as heigui (“black ghost”) and other terms. However, the basic framework and set of associations remained unchanged. This continuity is visible despite the fact that European regimes of bondage expanded into China in this period, presumably offering new ways to employ and describe African labour. Nonetheless, Ming and Qing sources provide a fuller picture of the tasks assigned to Africans than earlier sources. These included working in docks mending the hulls of damaged or grounded ships and guarding their masters’ persons and homes. Yet within this context pressing questions remain regarding Africans’ personal experiences and agency. These unfortunately are hard to reconstruct from the Sinographic evidence, as it stands today.