Introduction
The controversy surrounding the mui tsai was one of the most protracted and absorbing colonial policy disputes of the interwar years, in both the colonies of Hong Kong and Malaya, as well as in Britain. In spite of this, it is hardly remembered today. Those who were once mui tsai—domestic bonded servants, often children, always female—are today often silent out of shame. Few have come forward to give their stories, and those who do often insist on anonymity. The life stories of the mui tsai, if told, would fall along a spectrum ranging from stories of benign adoption and light household drudgery to the most horrific stories of abuse and exploitation. These stories unfolded within the walls of the household, and within the even more private walls of a young girl's fear; they enter history only with the utmost reluctance.
In the past 20 years, however, there has been a gratifying increase in scholarly interest and writing on the mui tsai in Hong Kong. Ex-mui tsai are increasingly treated with sympathy rather than with stigma; several have ventured to give named testimony to skilled and sympathetic oral historians,Footnote 1 and a handful have even produced autobiographies.Footnote 2 However, the field is still small. Secondary literature is scanty, and most of it derives still from a narrow corpus of primary and classic secondary writing. Almost all of it is focused on Hong Kong.Footnote 3 The case of Malaya has been largely sidelined in mui tsai history. It is my conviction, and it will be a central argument of this paper, that the Malaysian case sheds light on a hitherto unappreciated dimension to mui tsai studies: the place of racial considerations on the road towards the legal abolition of the mui tsai in 1939.
Most academic studies of the mui tsai system have focused on its exploitative nature, the inertia of the colonial state in acknowledging it as a social problem, and the role of British missionaries and activists, particularly women, in agitating for its abolition. Susan Pedersen has highlighted the central role of British imperial women in shaping colonial views and policies on the mui tsai, under a framework of action she terms ‘maternalism’. Maternalists, in her analysis, were educated, politically conscious and often wealthy women who sought to ‘represent and protect women and children. . .presumed to be less fortunate or more vulnerable’ out of a sense of maternal as well as Christian moral duty.Footnote 4 Other historiographical perspectives set the mui tsai issue in the contexts of social and labour history, feminist history, and the history of health and medicine.Footnote 5
Where scholarship has most consistently addressed the mui tsai as a racial question, it has tended to regard it as a case study in imperial and orientalist knowledge formation. Here, Karen Yuen's paper has provided a thorough analysis of what she sees as the racialized ‘Other-construction’ by both British officials and Chinese community leaders. Both parties, she points out, operated from highly conservative conceptions of gender and family, and the ‘transnational discourse on Chineseness’ that arose from their interactions yielded a body of knowledge about ‘the Chinese’ that ultimately justified the colonial government's bovine attitude to the abolition of the mui tsai.Footnote 6
In this paper I hope to show that the treatment of the mui tsai issue as an essentially Chinese problem yields an inadequate understanding of the controversy and its outcome, and sidelines important differences that existed between its occurrence in Hong Kong and Malaya.
In the first section I provide a brief history of the mui tsai controversy. The second section focuses on the slippery nature of the term ‘mui tsai’; here, I draw attention to the difficulties faced by colonial officials who, under increasing pressure from abolitionists and anti-slavery activists at home, struggled in vain to find a working definition for what constituted a mui tsai, in order that they might legislate effectively against it. This definitional problem is crucial to my subsequent argument: I show that the colonial state historically understood the mui tsai as a ‘Chinese question’, and that the slow crawl towards abolition can be understood as the process of the ‘official mind’ in Malaya coming to terms with the inadequacy of this definition. The third section, drawing on hitherto unused source material, then examines a pivotal moment in this process of reckoning: the Mui Tsai Commission to Hong Kong and Malaya of 1936. I end with some considerations of the significance of the Malayan mui tsai to wider mui tsai scholarship.
A brief history
The first death of the mui tsai question
In the autumn of 1870, Chief Justice Sir John Smale tried two cases concerning the sale of young female children in Hong Kong. The first case involved a girl who had been sold three times: first by her parents to a man who then transferred her to the care of a married couple for a $50 pledge. In this household, she was used as a servant. One day, the mistress of the household beat the girl, causing her to run away. She was promptly picked up on the streets by a second woman and sold to a brothel, where she was forced into prostitution. The second case involved five individuals, all complicit in kidnapping, purchasing and selling two young girls for ‘immoral purposes’.Footnote 7
On both counts, Justice Smale convicted the accused parties, and in doing so, drew public attention to what was by that time an alarmingly widespread traffic in young girls. He was, however, the first to frame the mui tsai problem in Hong Kong in terms of ‘Chinese slavery’, and indeed had a history of doing so: earlier in 1861 he had caused a great stir by condemning Chinese coolie emigration as a slave trade as well. In his opinion, slavery could never be considered exempt under the British Proclamation of 1841, which prescribed the classic British policy towards its colonies: that of governance by non-intervention in indigenous customs, insofar as this excluded ‘every description of torture’. When Justice Smale condemned the practise of buying and selling children for domestic servitude as a form of slavery and a ‘blot on the honour of England’, he was casting Chinese custom against a universal morality he perceived to be embodied by imperial rule.Footnote 8
Justice Smale's judgment met resistance from local Chinese elites, who found it thoroughly insensitive. They insisted that the mui tsai system had always been a form of charity. The girls who were sold, they argued, came from desperately poor families who could not afford to keep their ‘superfluous offspring’.Footnote 9 The system was perceived as one of mutual benefit, a better alternative to infanticide or starvation. The family benefited from the significant sum of money they would receive for the girl, and from being relieved of all future financial burden. The girl would be better fed and housed than she would have been in her destitute natal family, and traditionally she would be married off when she came of age at her benefactors’ expense, likely into a higher social class. And naturally, the benefactor family would benefit from cheap household labour. It was, in short, a win-win situation, or rather the best choice one could make between two evils of infanticide and selling one's child, both of which were necessitated by the fact of stark poverty throughout China.
Naturally, most of the ardent defenders of the system were Chinese elites, themselves keepers of mui tsai. It was eminently in their interests to maintain the status quo. After considerable debate, local Chinese merchants and gentry assembled and jointly petitioned the Hong Kong Governor, John Pope Hennessy, in 1878 to form an association to suppress kidnapping, abduction, trafficking and abuse. This amounted to treating the symptoms, rather than attacking the system, of mui tsai keeping. Out of this petition came the Hong Kong Po Leung Kuk, founded on 8 November that year to provide a place of refuge for abducted and mistreated women and girls. Two reports were produced for the Colonial Office and the Registrar General in 1879 and 1883 respectively, reiterating the premises of the petition and recommending that anti-trafficking laws be strengthened instead. Three years later, an identical institution was set up in Singapore under the aegis of the Protector of Chinese, shortly followed by several more in Kuala Lumpur, Ipoh, Malacca, Penang and Kedah.Footnote 10
The Po Leung Kuk, as much in Hong Kong as in the Straits Settlements, was really more than an institution; it represented a wholesale approach to moral taxonomy. As both H. J. Lethbridge and Elizabeth Sinn have pointed out, the Po Leung Kuk was designed not to eliminate the fact of transfer of women and girls in itself, but rather to distinguish between legitimate and illegitimate forms of transfer: between ‘good transfers’ like marriage, concubinage or domestic service and ‘bad transfers’ like abduction and prostitution; in short, to curb the abuse (unlawful) rather than the use (lawful) of the system.Footnote 11 In allying themselves with the Chinese elite defence of the mui tsai, some historians assert that the colonial powers were shoring up the patriarchal authority implicit in a system that conferred the right to dispose of one's daughters as one saw fit.Footnote 12 Furthermore, this alliance enabled the colonial state to claim that it was acting on the behalf of the Chinese as a whole. As Yuen puts it, it enabled them to ‘theorise Chineseness’,Footnote 13 and to use this knowledge to retrospectively justify the laissez-faire that in truth already inhered in the proclamation of 1841. In other words, the colonial authorities threw their lot behind the ‘politically useful fiction’ of a ‘representative Chinese view’.Footnote 14 In Singapore, then, it is no coincidence that the mui tsai problem, as it was beginning to be called, was assigned almost without question to the Straits Settlements’ Chinese Protectorate.
Thus, the outcry that had begun to amass in the wake of Justice Smale's verdicts, and which had anyway never enjoyed any significant purchase in the Straits Settlements, petered out. Mollified by the Po Leung Kuk and vague promises of vigilance, it lay dormant for 30 years. But the seeds of the controversy were all sown in this period, and the themes that arose from these debates—slavery, the oppression of women, the scope of imperial responsibility, Chinese elite intransigence, and ‘Chineseness’—would all be resurrected in 1919, when the question of the mui tsai was once again raised ingloriously in the House of Commons by Clara and Hugh Haslewood, who were both living in Hong Kong at the time. This time, the issue would not subside as easily.
Resurrection and combat
The Haslewoods were staunch Christians, and upon hearing a sermon at the Hong Kong Anglican cathedral on the widespread practice of keeping girl slaves in the colony, they promptly sailed for England in December 1919, determined to rouse the awareness and conscience of the English public. Clara Haslewood's energies derived from a long tradition of British feminist anti-slavery campaigning, which since the late nineteenth century provided a crucial avenue through which British women had begun to emerge in public life.Footnote 15 Together, Clara Haslewood and her husband launched a tireless siege against the mui tsai system in Hong Kong. She was particularly shrill in her charge that the government was condoning ‘slavery under the British flag’.Footnote 16 The Governor of Hong Kong, Sir Edward Stubbs, brushed off these charges as the fulminations of a person ‘well-known to be. . .of unbalanced mind’.Footnote 17 But he underestimated the Haslewoods, and the British public. Over the next four years, pressure mounted in the House of Commons. The issue, appropriated by the press, humanitarian lobbyists, religious groups, sympathetic MPs and women's organizations, provoked a deluge of parliamentary questions and letters of protest. It escalated in volume and indignation until, one evening in 1922, Winston Churchill finally snapped, brusquely declaring that he was ‘not prepared to go on defending this thing’ and that he did not ‘care a rap what the local consequences were’—the mui tsai must be declared free, he instructed, and ‘so far as administrative measures can make it, this freedom must be real’.Footnote 18
The very next year, on 15 February 1923, Churchill pushed through a Female Domestic Service Ordinance in Hong Kong's Legislative Council, which defined a mui tsai as an employed female domestic servant and declared that any money that passed in the transaction of a mui tsai did not confer any right on the persons acquiring her services, nor would any subsequent person be able to employ a mui tsai under ten years of age. The question was cast as one of labour and employment, rather than slavery. ‘Thus disappears a system which had been likened to slavery by its opponents and to adoption by its supporters’, the Review of Legislation for that year declared.Footnote 19 Similar legislation was passed in the Straits Settlements and the Federated Malay States in 1925,Footnote 20 and these proved to be completely successful in convincing the British public, as well as the various Hong Kong societies and individual campaigners, that their mission had been achieved, and the problem had been solved—at least, for the next six years.
There was, however, little attempt to enforce these lofty declarations. In the five years from the passing of the Ordinance in Hong Kong in 1923, not a single prosecution was initiated under it.Footnote 21 For the most part, colonial officials still clung stubbornly to the ‘representative Chinese view’ held by the elite merchant class in Hong Kong (with whom the British fostered keen and close ties due to the considerable economic influence they wielded in the Colony), even though there were many groups that were, even at this early stage, in favour of abolition.Footnote 22 In Malaya many officials were even of the opinion that the Female Domestic Servants Ordinance was superfluous, since in principle the Labour Ordinance of 1912 and the Ordinance for the Protection of Women and Girls in 1914 were more than adequate protective measures.Footnote 23
But the sterility of the legislation did not long escape the attention of lobbyist groups. In 1927, the revolutionary Chinese Nationalist government passed a decree abolishing slavery in China and announced that all mui tsai transaction deeds were henceforth cancelled. Taking this cue, the Anti-Mui Tsai Society in Hong Kong immediately began to clamour for a process of mui tsai registration to be implemented. In June 1929 Britain voted in a labour government which, propped up by wide liberal support, pledged itself to a new agenda of active trusteeship that had begun to dominate colonial rule since 1925. And by the beginning of the 1930s, British women, newly enfranchized, were throwing their weight behind the question with renewed vigour.
The failure of the eponymous Ordinance
The Mui Tsai Ordinance was finally passed in Hong Kong in 1930 and correspondingly in the Straits Settlements in 1932. It gave a much wider definition to ‘mui tsai’ than had its predecessor, the Female Domestic Servants Ordinance. It demanded registration of all existing mui tsai, and also prohibited further employment of mui tsai. Most absurdly, it also declared that mui tsai would simply cease to be mui tsai upon turning eighteen or getting married, whichever came first. The whole problem (so officials triumphantly reasoned) would then neatly die out of its own accord. The vanishing mui tsai problem would be reflected in the falling numbers of officially registered mui tsai, in a farcical itemization of ‘cancellations’; that is to say, the various ways in which girls were summarily expelled from their mui tsai status: ‘admitted to Po Lung Kuk’, ‘left the state’, ‘married’, ‘returned to parents’, ‘reached the age of eighteen’, and most curiously, ‘absconded’ and ‘miscellaneous’.Footnote 24
Perhaps not surprisingly, like its predecessor, the Mui Tsai Ordinance was a failure. The registration period was so brief, and the ordinance so poorly and confusingly gazetted, that the resultant registers six months later almost certainly did not reflect the actual magnitude of the problem. Stubbornly, the administration forged ahead, and there followed several years of assiduous inspections conducted by Lady Inspectors appointed by and reporting to the Chinese Protectorate. But it seemed that registration and inspection had the unfortunate effect of driving the system underground. Evasion was widespread, and owners and brokers took to using the term ‘adopted daughter’ or yeung lui (养女) to escape registration, both in their houses and at the immigration offices.
Thus, in the wake of the Mui Tsai Ordinance, quite contrary to its intended effect, problems concerning women and mui tsai steadily increased in importance, even as the numbers on the mui tsai registers obediently decreased year on year. In Malaya, where the mui tsai had previously been consigned to the ‘Notes’ section of the Monthly Reviews of Chinese Affairs, they were beginning to occupy whole sections on their own, listed second in importance only to ‘The Situation in China’, which always came first, and ‘the Kuomintang in Malaya’. By 1936 the problem had reached sufficient dimensions in Britain's colonial gaze as to warrant the deploying of a special commission to the Colonies.
A commission disunited
The Mui Tsai Commission, consisting of Wilfrid Woods, C. A. Willis and Edith Picton-Turbervill, was given the task of going both to Hong Kong and Malaya to gather evidence on the mui tsai problem. Several weeks before they were due to sail for the colonies, notices appeared in local newspapers appealing to local residents to volunteer information on the mui tsai. The Commission spent three weeks at various official locations, reading and recording written and oral testimonies by a wide range of participants: colonial officials and inspectors, local community leaders and civil servants, missionary women, social workers, owners of mui tsai and even several mui tsai themselves. But the Commission Report, dutifully issued in 1937, reflected a profound disunity of views despite being based on the same corpus of evidence. Woods and Willis issued a ‘Majority Report’, while Picton-Turbervill produced a ‘Minority Report’, which departed entirely from the views held by her male co-workers. The majority report declared that in its considered opinion, ‘schemes for the protection of women and children [were] both politically objectionable and administratively impracticable’ and it denied that there was evidence of ‘any widespread ill-treatment, cruelty or moral degradation of children by the Chinese in Hong Kong or Malaya’.Footnote 25 The minority report, wholly opposed to this rosy view, insisted that ‘notification, and where necessary inspection, of all transferred girls under twelve years of age’ was what was needed instead.Footnote 26 ‘That a transferred child stands in peril is abundantly clear,’ Picton-Turbervill wrote in the Minority Report, and concluded therefore that ‘there is a clamant necessity for a new and comprehensive ‘Children's Bill’ which should consolidate and amend where necessary the whole corpus of legislation for the protection of children in Malaya and Hong Kong’.Footnote 27
Sir George Maxwell, an energetic Malayan colonial administrator who had become interested in the mui tsai issue through his Vice-Chairmanship at the League of Nations Committee on Slavery, exerted substantial influence on Picton-Turbervill and the minority report. The problem, to him, was quite simply one of the ‘child in a stranger's home’, a situation which he thought inevitably engendered abuse. In short, both Maxwell and Picton-Turbervill wanted nothing less than a full redefinition of the problem of the mui tsai as one of child welfare.
It was to be this bold Minority Report that, after some squabbling and feet-dragging, was eventually adopted. And it was the Governor of the Straits Settlements, Shenton Thomas, who first accepted and agreed to implement its recommendations in Ordinance No. 17 of 1939, or what came to be known as the Child Act. Indeed, of the 53 Ordinances passed in the Straits Settlements that year, nearly half were concerned with social legislation, four of which were passed solely in the interest of children (Nos. 8, 17, 18 and 43). Ordinance No. 17, as per Picton-Turbervill's counsel, provided for the employment, neglect and ill-treatment of children (defined as persons under the age of 14), and crucially, made mandatory the registration of all transfers of such persons away from their natal home. It also vested in the Protector the power to inspect a child should he deem there to be some reason to believe it was being abused or mistreated. Ordinance No. 18, for the first time in the history of the colony, prescribed legal procedures by which children could be adopted.
Understanding the disunity of the Commission Report, and how and why the Colonial Office finally went the way it did, formed a large part of the impulse behind this paper. The next section looks more closely at the perils of definition that plagued colonial understanding of the mui tsai question, which will go some way towards explaining the successive failures of the legislation outlined above.
Definitions
The mui tsai problem was essentially one of definition. The term mui tsai is a Cantonese rendering of mei-zi,Footnote 28 which roughly translates as ‘little sister’—a mild, innocuous label—and it was an ambiguity that plagued attempts to regulate the system. What is clear, though, is that it began as a question of slavery within the Chinese community.Footnote 29 Slavery dressed the problem in moral terms, easily assimilated by a British public whose recent history had seen over 100 years of anti-slavery campaigning. Later on, many British feminists, newly enfranchized and becoming increasingly engaged with colonial policy, were keen to cast the mui tsai problem as a ‘women's issue’ concerning oppressed Chinese women and girls at the combined mercy of British and Chinese men.Footnote 30 Chinese and other implicated elites, as we have seen, claimed the system as an age-old custom. But none of these—slavery, female oppression, tradition—were sufficient for the administration to properly identify the mui tsai. The term itself proved to be extraordinarily slippery.
Mui tsai were understood to be young, female household drudges, but the system has variously been called ‘adoption’, ‘concubinage’, ‘betrothal’, ‘prostitution’ and ‘slavery’. This uncertainty arose from the complex patriarchy that characterized traditional Chinese family structures. Igor Kopytoff and Suzanne Miers distinguished between two modes of belonging in kin groups: family members either ‘belong to’ or ‘belong in’ a given kin group.Footnote 31 James Watson has applied this analysis to Chinese women who, he claims, regardless of status or class, inevitably ‘belonged to’ the traditional family unit, while Chinese men ‘belonged in’ it. But the distinction was not simply a matter of gender; it also reflected on the exteriority of kinship. Watson further suggests that legitimate, birthright sons ‘belong in’ the family, while adopted sons, having been literally acquired from outside the family unit, ‘belonged to’ it.Footnote 32 The repercussions for mui tsai are clear: as both female and exterior to the family unit they ‘belonged to’, they were in a sense twice-removed from legitimacy. As one woman testified to the Commission, ‘I do not treat [my mui tsai] exactly as a daughter because it would be against family tradition’.Footnote 33 It is to some extent this attitude that fuelled allegations that the mui tsai relationship placed girls in positions more liable to abuse than so-called ‘natural’ relationships.Footnote 34
There were many other forms of exploitation within the traditional Chinese family for women who came from outside the natal family. One of the more studied systems is that of child betrothals, known as the san po tsai Footnote 35 system, which involved the transfer of young (at times even infant) girls from their natal families to the households of pre-arranged future husbands, who were at the time of transfer often as young or even younger than the girls. The motive was economy, as the businessman and prominent Singaporean Straits Chinese leader, Lee Kong Chian, revealed: ‘the father-in-law gets the girl's labour for her keep without wages, and the ceremony on puberty is much cheaper than a [normal] wedding would be’.Footnote 36 As with the mui tsai, the san po tsai system was a kind of mutually beneficial redistribution of female children: the girl's family would be relieved of the expenses of raising her, and the wealthier boy's family would gain a cheap bride and free labour. From the outside, the two systems were virtually indistinguishable. As one commentator wrote, ‘It would be hard for you or me to say where a daughter-in-law's tasks begin and where a mui-tsai's end’.Footnote 37
As Kopytoff and Miers have pointed out in the case of Africa, neither the criteria of property nor of saleability are useful in separating ‘slavery’ from ‘kinship’ in societies where rights in wives, children and other kin-group members are acquired through financial transfer or barter, in ways that seem to imply propertied ownership, at least by ‘Western’ standards.Footnote 38 Rather, they say, there exists a ‘slavery-to-kinship continuum’ in which the nominal categories of role and status within a household segue indistinctly into one another.Footnote 39 In Chinese family structures, there were similarly blurred boundaries and complex relationships between wives, second wives, concubines, mui tsai and daughters; and there was typically monetary consideration accompanying their transfer between largely patrilineal family units.
The mui tsai question conflated these issues, and brought them all uncomfortably to the fore. The public outrage whipped up by the Haslewoods’ activism began to find its way, in the form of pointed questions, to the House of Commons. Stock official responses ranged from flustered admission to denial. When the question was first raised in parliament in March 1920, Lieutenant-Colonel Amery was caught off-guard and pressed into a feeble response that there existed not slavery but ‘another practice, and I do not remember the Chinese term for it, which is the adoption of girls and youths for domestic service, in which case a lump sum is paid to the parent or guardian’.Footnote 40 Later that same year, lobbiers found him better prepared: ‘Slavery,’ he insisted, ‘does not exist in Hong Kong or any other British Colony’.Footnote 41
There was in fact legal justification for this claim: under British law, monied transactions were not recognized as conferring any right of ownership onto the purchaser when the ‘property’ in question was a human being. In purely legislative terms, then, slavery indeed did not exist. But that the law did not recognize it did not itself eradicate the problem. Money did pass with bodies, whether or not British law assented to what rights that money conferred. The League of Nations Committee of Experts on Slavery more simply advised that in the case of human beings, ‘the courts should determine if, in fact, the transaction is one of purchase and sale and, if so, to deal with it as an act of enslavement’.Footnote 42 This seemed straightforward enough. If money accompanied the transfer of a mui tsai, this was proof of sale; sale was proof of the transfer of property, and property implied enslavement. Therefore, a mui tsai was a slave.
However, this line of reasoning was thoroughly confounded when it was pointed out, almost from the start, that money accompanied all forms of female mobility in Chinese society. Marriage was accompanied by peng kam, a form of dowry that passed from the bridegroom's family to the bride's family, and was sometimes used to purchase the bride's trousseau or even returned to the groom's family in a show of good faith. Adoption, even of sons, was generally accompanied by a red packet of ceremonial money. Concubines were simply bought outright, and generally commanded higher prices than kit fat, or principal, wives. Prostitution, of course, was already a highly monetized trade in sex. Then there were the mui tsai; for these transactions, so the claim went, Malayan Chinese distinguished between mai (buy) and sung (give-as-present)—the former for mui tsai sale and the latter as transferring into adoption. Most confusingly, as one official complained, ‘money passes in both cases’.Footnote 43 Clearly, the simple categories of ‘purchase’ and ‘sale’, insofar as these were determined by the transfer of money, were not applicable. However, colonial officials well into the 1930s remained—and were repeatedly criticized for being—‘blinded by the money transactions’ that characterized Chinese kinship practices.Footnote 44
The mui tsai thus resisted the colonial state's best attempts at definition, and processes put into place to register them, as we have seen in the previous section, largely failed. The Secretary of Chinese Affairs in Malaya, upon being questioned by the Mui Tsai Commission on why exactly it was so difficult to ‘catch’ unregistered mui tsai, replied wearily: ‘Because hard and fast rules cannot be drawn with the Chinese population of Malaya as in England’. He sketched the following scenario: A coffee-shopkeeper from Hainan marries a Hainanese woman as a principal wife. Upon emigrating to Malaya to work, he marries a Cantonese woman and has children. He then returns to China and usually will not take her with him. Her only choice is to ‘follow’Footnote 45 another man, who will usually not want her children. Thus, ‘she has to find some other place for them to live, where they will go as mui tsai or female domestic servants or as adopted daughters or as semi-adopted or quasi-adopted daughters or as prospective daughters-in-law. Their future is in all cases uncertain’, he finished morosely, adding that there were scores of such happenings every day.Footnote 46 George Maxwell, familiar with many of these ‘scores of such happenings’, was perhaps most keenly aware of this problem of definition. He declared in his memorandum on the mui tsai that ‘until someone can define a “mui-tsai”, and distinguish her from a “foster daughter”, a “quasi-adopted daughter” and a “prospective daughter-in-law”, no one will be able to offer any opinion of any value upon the present number of unregistered mui-tsai in Hong Kong (and Malaya) today. That, in a few words, is the brutal truth’.Footnote 47
A final problem that plagued colonial attempts to define a mui tsai was the fluidity of their status. The women were not always mui tsai, nor might they always remain so. A girl might be presented to another family as a san po tsai, but if the betrothed son decided to reject the marriage when the time came, she would likely stay on in the family as a household drudge. Conversely, a mui tsai might later be adopted as a concubine and thus establish herself in the household as a legitimate matriarch. An autobiography by Shirley Lim, who grew up in Malacca, tells the story of just such a case, in which her own position in her household, as a legitimate daughter, was supplanted by a mui tsai who seduced and married her father.Footnote 48 But marriage was not always a release. Koh Choo Chin stated that in her experience as a social worker and assistant Lady Protector of Mui Tsai, many simply became subservient to a domineering mother-in-law instead.Footnote 49 Jaschok also relates a situation in which a mui tsai altered her status simply by staying in the family long enough: ‘with occasional money-gifts, [the mui tsai's status] was subtly changed to that of an amah’[servant].Footnote 50 Another social worker testified that she had known of several instances in which girls had been adopted into a childless family, but into which a natural child was subsequently born. ‘The adopted girl has then reverted to the position of mui tsai’, she said.Footnote 51
What seems clear from these examples is that the mui tsai, in ‘belonging to’ the kin group, had no recourse to alter their own statuses; in other words, they did not define themselves, but always were defined. As Jaschok observes, ‘those who granted that recognition had to be members of the dominant structures of society, namely men’.Footnote 52 Jaschok had in mind the men to whom women and girls had to attach themselves in society and within the family unit; but the group of ‘men’ to whom the definition of the mui tsai was most important was, naturally, the colonial administration. Definitions were essential when it came to the mui tsai: in order for officials to deny, inspect, register or abolish the system, they needed to know what exactly the system was.
A trans-racial problem
In the face of the confusion in which definitions of the mui tsai seemed irretrievably mired, colonial officials took refuge in the knowledge that, at the very least, it was most definitely a Chinese problem. But was it? I turn now to the question of the ‘Chineseness’ of the mui tsai question, which, as I have suggested earlier, remains very much taken for granted in most of the secondary literature on the mui tsai.Footnote 53
Before she left for the Colonies, Miss Picton-Turbervill had lamented to her fellow Commissioners that ‘anything can happen in a Chinese house because the people are so secretive. The houses are very much walled-in and there are back quarters which one cannot really penetrate’.Footnote 54 She was, of course, referring to the difficulties encountered by Lady Inspectors of the Mui Tsai in ‘penetrating’ the houses to be inspected. But the comment exhibits a wider tendency on the part of the administration to cast the mui tsai problem as an ‘essentially Chinese’ one, and to make statements about ‘the Chinese’ as a homogenous group to which one attached qualities like ‘secrecy’ or ‘impenetrability’ as markers of their cultural difference. Sweeping statements about ‘the Chinese’ frequently found currency with members of the British public. A letter forwarded to the Haslewoods on 18 May 1937, which caused them great exasperation, asserted:
The [mui tsai] system is really no more than a system of adoption. All Chinese girls whether mui tsai or not work hard all day long and if they do become prostitutes, what of it? For Chinese and Asiatic girls take up prostitution as a career just as our girls take up service or nursing and think nothing of it. It is just another case of the eastern point of view being different from the western and it is a great presumption to think that the western view is always right.Footnote 55
Chinese adoption was still seen as a ‘vague and shadowy term’, at least according to the Chairman of the Commission, who added that he did not know if it was vague and shadowy ‘in the minds of the Chinese; I am afraid it is in my mind, because I do not know enough of Chinese domestic life to say’.Footnote 56 The Secretary for Chinese Affairs in Malaya, when interviewed by the Commission, suggested that mui tsai status was inheritable, ‘simply because Chinese psychology accepts that situation’.Footnote 57 The Commission's knowledge-gathering on the mui tsai, on these curious terms, seemed almost like a brave sortie into the ‘vague and shadowy’ backrooms and inner workings of the ‘Chinese mind’.
There were problems in essentializing ‘the Chinese’ in this way, and they came importantly to the fore over the course of the mui tsai issue, which was by default viewed as a problem peculiar to a particular race. The establishment of the ‘Chinese Protectorate’ represents an early manifestation of this tendency: a single administrative entity for what was perceived to be a single cultural entity. Within the Federated Malay States and the Settlements, the Chinese Protectorate was designed to be the first port of call for all Chinese settlers to voice their troubles to Chinese-speaking British officials. Ruling by divide in this way, there was little need for Chinese community leaders to interact with the Malay chiefs and rulers; it was thus easier to keep the two ethnic groups culturally essentialized and discrete, even those of mixed descent. Outside these Protected states, however, ethnic boundaries were far more porous; Barbara Andaya and Leonard Andaya point to Johor as an example of Chinese-Malay administrative relations that far better reflected the fluid and complex composition of Malayan society than simplistic colonial attitudes allowed for.Footnote 58 Unfortunately, it was just these simplistic attitudes that continued to inform the British handling of the mui tsai issue as an essentially Chinese one.
The essentialization of ‘the Chinese’ with respect to the mui tsai problem has also had historiographical consequences. Historians have similarly tended to treat it as a moment of Chinese modernity, of challenge to prevailing Chinese traditions and norms, and of the liberation of women, placing the problem generally within the boundaries of ‘Chinese’, and particularly focusing on Hong Kong. Karen Yuen is one of few voices championing a transnational approach, in which the mui tsai controversy becomes an instance of knowledge construction of, as well as by, the Chinese; but, as indicated earlier, Yuen, too, operates within the unquestioned category of ‘Chineseness’ even whilst questioning the national categories of ‘Hong Kong’ and ‘Malaya’. While undoubtedly, as she says, colonial definitions of ‘Chineseness’ ‘were not restricted by the geographical boundaries of the Chinese nation-state’ and that the mui tsai issue must be understood as a custom carried along with the Chinese diaspora,Footnote 59 we cannot ignore the concrete differences that existed between Hong Kong and Malaya. I argue that while colonial authorities did indeed ‘transnationally’ perceive the mui tsai as a ‘particular and essentially Chinese practice’, this perception was subsequently confounded by particular local conditions that Yuen's rubric is disposed to preclude. In this section I wish to stress the particular ‘trans-racial’ dimensions of the Malayan case as it began to appear to the colonial state.
Because the mui tsai had been defined as a solely Chinese problem, there had been little attempt to understand the issue outside the context of the Chinese community. The Secretary for Chinese Affairs in Malaya in 1936, A. B. Jordan, reported to the Commission that ‘wherever you have a considerable Chinese population, then the status of mui tsai will be created’.Footnote 60 In 1934 after the adoption of the Mui Tsai Ordinance, the Chinese Protectorate deemed it necessary to appoint Lady Inspectors of Mui Tsai in Singapore and Penang, but none at all for the Malay States: a fact George Maxwell objected to, pointing out that if there were two Lady Inspectors for the 589 Singaporean mui tsai and the 492 Penangite mui tsai respectively, then there should surely be one for the 987 mui tsai in the Federated Malay States—not to mention the numerous unregistered mui tsai in the Unfederated Malay States, whom Maxwell was certain existed despite the failure of the Register to reflect this.Footnote 61 Indeed, the Chinese Protectors in the various Malay States were also embroiled in the mui tsai problem, with one complaining that upon the gazetting of registration in 1933, he had been inundated with anxious applicants who appeared to own ‘mui tsai’, 110 of which were eventually registered. Mr Williams, Protector of Chinese in Kedah, opined that this number was too high in relation to the small Chinese population in Perlis. Upon further investigation by the Lady Inspector of Mui Tsai in Penang, 91 of these were declared to have been registered under a mistaken notion of what a mui tsai was. The register was subsequently purged to 19. ‘There is no one in the State competent to decide who [are or are not] mui tsai’, O. R. Howitt, the British Adviser in Perlis, finally complained.Footnote 62
That colonial officials were quick to purge the register in accordance with ‘racially proportionate’ results which demonstrates the distinctly racialized nature of the evidence taken in Malaya in comparison with Hong Kong. A close examination of the notes gathered by the Commissioners reveals significant differences between the testimonials of the mui tsai issue in Malaya and those of Hong Kong. In particular, the questionnaires set by the Commission, and the answers given by the respondents, both reflect concerns of a multi-racial nature regarding the mui tsai, which few colonial officials (and indeed historians) picked up on, and which simply did not exist in Hong Kong. The Malaya questionnaire contains four specifically racial questions that do not appear in the Hong Kong questionnaire, inquiring after various permutations of inter-ethnic mui tsai employment: for example, whether there were non-Chinese women or children employed as mui tsai or in systems of bondage analogous to that of a mui tsai; or whether there were Chinese women or children employed as mui tsai by non-Chinese families; in other words, whether anyone owned or knew of someone who owned non-Chinese mui tsai. Footnote 63 The answers obtained by the Commission in response revealed a bewilderingly plural dimension to the problem of the mui tsai.
It turned out that dubious adoptions were not confined to the Chinese community.Footnote 64 There seemed to exist comparable systems of domestic servitude that involved neither Chinese employers nor Chinese girls. The Lady Inspector of Mui Tsai in Singapore gave an example of a Malay girl employed (and ill-treated) by an Arab family, though she stated she had never heard of a mui tsai system amongst the Malays.Footnote 65 The Acting Secretary for Chinese Affairs at the time, J. M. Black, elsewhere suggested that ‘bona fide’ adoption, rather than employment, was more common among the Malays.Footnote 66 There was also substantial evidence for the existence of an analogous system of domestic female labour amongst the Indian community, particularly between castes (e.g. the Jaffna Tamils and the Hindus), though there were allegedly few cases of abuse, and of those, fewer still that failed to reach sympathetic or official ears.Footnote 67
The Chinese community in Malaya also took non-Chinese girls into their houses. This appears to have been extremely commonplace and, I would suggest, more so in the 1930s than previously, as the Chinese population in Malaya began to put down roots, and as measures by the colonial government taken to curb female traffic began to take their toll on the supply of domestic labour from China.Footnote 68 Barron knew of Tamil, ‘mainly pariah’ girls, brought into Chinese families as domestic servants comparable with mui tsai, and cited nine cases he was personally acquainted with in Penang.Footnote 69 Black testified that there were Siamese and Tamil girls who had been registered as mui tsai in Chinese families, with the Siamese being more common in North Malaya; this is corroborated independently by several witnesses who all personally knew of several such cases in Kedah.Footnote 70 Indeed, on the Kedah register there were not only eight Siamese girls, but also three Tamil girls and three other ‘Chinese-Siamese’ girls, all residing in Chinese homes. Other evidence suggests there were considerably more than this: several subjects independently testified to the normalcy of Chinese families acquiring Tamil, Siamese, Javanese, Malay and ‘Annamite’ girls,Footnote 71 and there is an interesting suggestion by another that Siamese parents were also voluntarily selling their children to Malayan Chinese families.Footnote 72
Far less frequently, Chinese girls were sold into non-Chinese homes. An English missionary woman had heard of several such cases during her 14 years in Malacca, including that of a Chinese girl who had been ‘legally adopted’Footnote 73 into an Indian family, and who, as far as she knew, had not been ill-treated or co-opted into servitude.Footnote 74 These occurrences seem few and far between; however, it was noted that the purchase of Chinese children by Malay households was relatively common practice, due to a prevailing belief that ‘Chinese blood is superior to [that of the Malays] and that it makes a good addition to [Malay] blood’. Furthermore, he continued, the Malays liked ‘pale-faced Chinese girls’, and if the girl turned out to be ‘a good one, she would be regarded as an adopted daughter; if not, she would be regarded as a mui tsai’.Footnote 75 Finally, there had also been the case of one hapless Malay Chief in Ipoh, who, seemingly unconcerned about whether his girls were pale-faced or not, had been found and convicted of being in possession of nine girls—‘some Chinese, some Malay and some Indian’—all of whom were subsequently deemed ‘unregistered mui tsai’ and confiscated to the Po Leung Kuk.Footnote 76
Even where this trade did concern Chinese girls in Chinese homes, there were other complications which proved peculiar to Malaya. One such problem had been raised by Choo Kia Peng in a memorandum on the mui tsai and adoption in relation to slave trade. While noting that Malayan and Straits immigration controls were now sufficiently rigorous to deal with the traffic in mui tsai, he pointed out that there also now existed second and third generations of Chinese children indigenous to Malaya, of which he identified three types: children of Straits-born Chinese, children of China-born towkays (businessmen), and children of artisans and labourers.Footnote 77 Earlier colonial officials had pointed out that the Malayan mui tsai were different from Hong Kong mui tsai in at least one crucial way: they were much farther removed from their natal families. Unlike in Hong Kong, then, there was a growing set of indigenous Chinese children in Malaya, a fraction of whom might conceivably be sold, transferred or simply inherit their status as mui tsai, adopted daughters, daughters-in-law or any of the other definitional classes. The magnitude of the problem in Malaya, and the need for action, it seems, was even greater now that it was no longer simply an inflow that might be stemmed at the immigration ports, but was rather becoming something wholly indigenous. Furthermore, this aspect of the problem was also not restricted only to the Chinese. For example, evidence supplied by J. Handy, Assistant Controller of Labour in Singapore, suggested that as the Indian sojourner population became more rooted in Malaya, they increasingly substituted the Tamil or Ceylonese girls they had brought with them with Chinese amahs or mui tsai, whom he claimed were regarded and treated not as victims of sale, but as waged workers.Footnote 78 This complicated matters further, particularly in terms of immigration: for example, were customs officials to detain dubious Chinese children at the ports, but allow children and families of other races through without question?
Malaya, indeed, eventually proved to be a melting pot of child welfare problems. Judith Djamour's 1952 anthropological study of ‘adoption amongst Singaporean Malaysians’ examined the different and minutely gradated forms of adoption that existed in Singapore at the time. She distinguishes between anak angkat and anak beli (‘adopted’ and ‘bought’ children); of the former category, there are further categories of ‘nominal’, ‘foster’, ‘temporary’ and ‘total, unconditional’ adoptions; of the latter, Djamour consigns the practice in general (although she notes, not exclusively) to Chinese children sold into MalayFootnote 79 homes. In both, she identifies an element of monied transaction, and suggests that:
among wealthier Malay[s] a female anak beli is sometimes used as a domestic servant when she grows up . . . This practice is very rare and is frowned upon by the community, who pride themselves quite justifiably on being fairer in their treatment of anak beli than are the Chinese or Arab sections of the population, [both of which] have a long tradition of buying Chinese infant girls who are destined to become unpaid domestic servants as soon as they are old enough to be useful.Footnote 80
From this, we can see that roughly analogous systems of domestic servitude existed among Malay, Chinese and Arab populations in Malaya; that the practise of child transfer was much more prevalent than the colonial officials had likely previously thought; and that the problem, insofar as it had been defined in existing terms in legislation, was in no way confined solely to the Chinese community in Malaya.
In the light of all this evidence, George Maxwell perhaps put it best when, in great exasperation, he concluded that ‘it seems absurd to deal with the children of all these nationalities as ‘mui tsai’.Footnote 81 However fluid and confusing Chinese kinship relations were, and however difficult it had been in Hong Kong to agree on ‘where a san po tsai ended and a mui tsai began’, colonial officials knew the problem as a Chinese peculiarity; they could coherently delegate it to the Secretariat of Chinese Affairs; they could, as Yuen has indicated, construct knowledge about the Chinese from this issue. But in Malaya, faced with a slew of untidy, extra-racial counter-examples that resisted even this level of classification, it is clear why there was increasing contention over the question of whether a girl other than a Chinese could be a mui tsai.Footnote 82 The very phrasing of the problem here betrays the failure of colonial nomenclature. By the time the Commission was writing up its disunited reports, it was clear that the travails of taxonomy, which officials had thought consisted solely (though no less intractably) of distinguishing between wives, concubines, san po tsai, mui tsai, and adopted daughters were, in Malaya, only the tip of the proverbial iceberg.
Accounting for Malaya
As I have hoped to show in this paper, the over-emphasis on Hong Kong as the centre of mui tsai scholarship has tended to obscure interesting and unique differences between mui tsai cases in Malaya and Hong Kong. I suggest that the racial pluralism—what I call ‘trans-racial’ dimensions of the Malayan case—really helped to bring home to British colonial officials in Malaya the wider problem of child welfare. Since there appeared to be, most confusingly, Tamil, Siamese, Burmese and Malay ‘mui tsai’, what officials had called the ‘mui tsai problem’ was clearly not only a Chinese one, and had to be tackled on a wider scale: as a problem of child welfare and illegal adoption. J. D. Vaughan noted, as early as 1879, that ‘there is scarcely a family in the Straits that has not adopted children within its circle’.Footnote 83 Rosemary Firth found from her fieldwork in a single Malay rural community in Trengganu in 1942 that 53 out of a total number of 465 children (about one in eight) were being cared for by people other than their natural parents.Footnote 84 Judith Djamour found in a single fishing village in Singapore 1946 that there were 18 adopted children in 16 out of the 48 households, or fully one in five children not living in their natural households.Footnote 85 Cases like these would almost certainly have come under scrutiny by the increasingly paranoid British administration. None of these were formalized in any sense; not all of them were Chinese, domestic servants or even girls. What the investigations into the mui tsai issue in Malaya really did, therefore, was expose the extent to which child transfers were taking place at all class levels, in all communities, amongst all racial groups.
This shift towards a child-centred view of the mui tsai question was a local expression of a more general ideological trend. David Pomfret has recently drawn attention to this, arguing that over the course of the 1920s and 1930s, changing ideas about age, not gender, began to provide the primary impetus to mui tsai reform. ‘Commitments to the protection of [an idealized] childhood’, he writes, ‘had emerged at the forefront of liberal internationalist efforts to reshape the postwar order’. As early as the 1920s, demands for imperial reform were often made ‘in the name of the child’, acquiring additional weight from the mid-1920s as international organizations like the Save the Child Union (1920) and the League of Nations came to incorporate these anxieties about childhood into their agendas. Edith Picton-Turberville's Minority Report, Pomfret thus suggests, was ‘only the latest in a series of challenges throughout the period, in which those who strove for anti-slavery reform consistently identified age as a key ground upon which to stake out their claims’.Footnote 86 Yet if this rhetoric of childhood was in popular circulation from the 1920s, it only began to see concerted legislative expression in matters concerning the mui tsai towards the end of the 1930s. In this arena, the Straits Settlements government was quicker to act than its Hong Kong counterpart. It accepted the Minority Report more quickly than Hong Kong, and despite having lagged behind Hong Kong in mui tsai legislation throughout the 1930s (a fact which George Maxwell relentlessly brought to the attention of the Colonial Office),Footnote 87 it was quicker to finally abolish the mui tsai, at least in legislative terms, under the Children's Ordinance. I suggest that the circumstances which prevailed in the Straits Settlements (described above as the de-racialization of the mui tsai question) have not been taken fully into account by previous scholars, and should be weighted amongst the reasons for the swiftness and willingness with which government accepted the Minority Report, and proceeded to put the appropriate legislation into motion.
The impact of the Straits Settlements’ decision on Hong Kong has been mentioned, though not explored, in past mui tsai scholarship. Pomfret, whose work on the mui tsai focuses on Hong Kong and Saigon, accepts that the responses of other Crown Colonies to the mui tsai problem (notably in the Straits Settlements) ‘had been undertaken with greater zeal’. Miners too has pointed out that Malaya's acceptance of the Minority Report undoubtedly influenced the British administration in Hong Kong to accede to the same, since ‘it would be difficult for the Secretary of State to defend the adoption of divergent policies by two colonies in dealing with the same problem’.Footnote 88 That it was in fact the same problem was a relatively new realization with recent ideological roots. In 1925, a new British department of state, the Dominions Office, was formed as an entity entirely separate from the Colonial Office, with its own staff, directives and constitutional status. This reorganization of the Colonial Office enabled a more internally consistent approach to the Colonial Empire ‘proper’, and stimulated a growing recognition that many problems particular to individual colonies, including questions of intelligence, education and welfare, were in fact similar and should be resolved uniformly.Footnote 89
The pressure on Hong Kong to adopt the Minority Report once Malaya had already done so, its own internal pressures notwithstanding,Footnote 90 certainly arose in part from this new emphasis post-1925 on intra-empire uniformity. When news came through of the Straits Settlements’ acceptance of the Minority Report, George Maxwell dispatched a triumphant note to Sir John Harris, Secretary of the Anti-Slavery Society: ‘I have not the slightest doubt that the final decision [of the Hong Kong government] will also be in favour of the Minority Report . . . we have an overwhelming triumph’.Footnote 91 Whilst I do not argue here for a fully causal connection between the decisions of the Straits Settlements and Hong Kong governments, Maxwell's allusion to this intra-imperial relationship is highly suggestive, and has been obscured by the emphasis of present mui tsai scholarship on Hong Kong and China. More research is needed to illuminate these murky intra-imperial networks, and taking scholarship in this direction will require a more concerted attempt in the literature to take the case of the Malayan mui tsai fully into account.
To this end, I have sought here to begin by shedding light on some of the internal circumstances shaping the imperial treatment of the mui tsai in British Malaya. I suggest that in seeking to understand the complexities of mui tsai abolition, in particular the shift towards regarding it as a matter of child welfare, historians and scholars of the mui tsai might gain new insights from examining the uniquely trans-racial dimensions of the Malayan case. In Malaya, the mui tsai controversy persistently defied the colonial rhetoric of ethnic governance, which foundered in the face of the confusing plurality of the Malayan mui tsai. They could no longer be governed simply as a product of Chinese social vice; the mui tsai were not merely a ‘Chinese problem’. In recognizing this, Malayan officials sought to widen the terms under which the mui tsai could at last be abolished, and in doing so, lent important weight towards the impetus, within Malaya as well as in other parts of the empire, to abolish the mui tsai ‘in the name of the child’. For by the time the Minority Report arrived on his desk in 1937, the Governor of the Straits Settlements, Shenton Thomas, was probably already sympathetic to ‘the duty of a civilized Government to protect its children’Footnote 92—not simply Chinese girls arbitrarily dubbed ‘mui tsai’, but children of all races, shifting and slipping in and out of the households of this strange plural society under his care.
