version finale (aoыt 2011) SLAVE THAT ARE NOT SLAVES, YET REALLY ARE1 Alain Testart Any evaluation of slavery outside the Western world offers the following difficulty: while the opposition between freedom and dependency has been clear and evident in our law and in our political thinking since the Greeks, in southeast Asia
there is a whole gradation between complete dependency (the slave) and complete freedom. Even supposing that the category of slave existed in Asia, liberty does not appear as a major concept of Eastern political thought. In the West, the clear conceptual opposition is reinforced by the fact that the slave is always an outsider. The slave already lay outside the Aristotelian conception that recognised that the Greeks from one city could enslave other Greeks from another city when war was waged between them, but held that they became slaves only by accident, since barbarians--all those who were not Greek--were characterised as slaves by nature. The slave became more frankly an alien when the generic term designating him shifted during the Middle Ages from servus to sclavus or slavus--which gives us "slave"--from the name for the ethnic group (the Slavs), from which the Byzantine Empire and Venice recruited most of their servile contingents. This shift culminated in modern times, at the end of the sixteenth century
, when black people were used as slave labour for the exploitation of colonies in America. At that point, everything opposed the slave to the free man: his legal status
as non-free and non-citizen (or non-subject of a kingdom), his paganism (because the Christian West accepted the enslavement only of infidels or pagans), and finally, the colour of his skin. 1 This short introduction summarises theories and facts examined at length in my book L'Esclave, la Dette et le Pouvoir (Paris: Errance, 2001). Another article was devoted to the situation in Southeast Asia (Testart 2000).
2 There was nothing of the kind in the Far East, either in the Chinese Middle Empire or in Southeast Asia, where slavery for debt was often legal and widely practiced, right up to colonisation. Norodom, the King of Cambodia, on the eve of colonisation in 1884, told representatives of France that "Enslavement for debt is one of the foundations of the Cambodian state."2 Sometimes it was not legal, but there existed multiple forms of servitude that similarly allowed the reduction to dependency of one's kin, neighbours, and fellow people. Poverty led the poorest to sell themselves into servitude, after having sold their wives and children. The status of slave did carry some advantages that may seem astonishing to us: being dependant only on their masters, slaves escaped taxes and military service, which were often very onerous. Thus dependency could be preferable to a mere semblance of freedom. What liberty could there be for subjects crushed by taxes and forced into periods of service that were so long that Orientalism adopted the habit of referring to them as "serfs" forced into conscripted labor (corvйe)? For many, the choice seemed to be merely between one form of dependence and another. Thus everything was blurred in this Oriental Asia. Our first task is to sort through this multitude of statuses of dependency. But first, what is a slave? This question haunts the texts that we are going to read, those by the anonymous author of "Note sur l'Esclavage" (1863), Briffaut (1907) and Jules Silvestre (1880), as it had already haunted Andrй Baudrit's Human Cattle (1942). We read in his book, citing Raoul Postel, that a certain kind of slavery, for debt, "is not properly speaking a form of slavery" (Baudrit, 1942: 151). And similarly, Silvestre says about the indebted that "this is not really slavery" (Silvestre, 1880: 232). And then we hear Briffaut (1907: 275) speak of "domestic serfs" who seemed to be just servants engaged for the long term, perhaps in perpetuity. In fact, nothing justifies calling them "serfs" because there does not exist in Annamite or Chinese law a category that one might identify as serfdom. To read these texts, slavery is everywhere, but nowhere is it defined, nowhere is it identifiable. All these observers nevertheless understood that the situation in the Orient was very different from what has reigned in Europe of their day. Scholars like Briffaut 2 Reported by Paul Doumer, L'Indochine Franзaise. (Paris: Librairie Vuibert, 1930) (p. 238), quoted by
3 and Baudrit, and colonial administrators like Silvestre, Dartiguenave and Landes saw the importance of dependency, but their sole error--frequent in the writings of writers of that period, so that we cannot seriously reproach them--was to use the label "slavery" indiscriminately. They did so as men of the nineteenth century, careful to condemn any dependence by branding it with this shameful term. It suffices to cite the innumerable pamphlets and political writings of the day that denounced the condition of women in Europe as being that of slaves, a simple metaphor that is very telling about mentalities. Finally, if Baudrit, Silvestre, the anonymous author, and Briffaut used the slave label so liberally, it may be because they were French, and the French language does not possess terms like the English dualism between bondage and slavery.3 A slave is a dependant. Not every dependant is a slave, but any slave is certainly a dependant. So the first thing to say about a slave is that he or she is a dependant.4 And what is a dependant? A person who cannot, by him or herself, escape from a situation. In other words, because any social situation implies obligations, it is a person who, while having acquitted himself of his obligations, still finds himself burdened with the same obligations. A slave, after having done what the master demanded, still has to obey him. In vain does he perform everything the master asks, for he still remains a slave: this is what it means to be a dependant. A citizen must pay his taxes and once he has paid them, he must pay them again the next year. In a sense it is futile to pay each time what is demanded, for he will again the next year be burdened with the same obligation to pay. This is because the citizen is, on the issue of taxes (as that of military service among other things), dependent on the state. A waged worker, on the other hand, once he has performed the task for which he was hired, owes nothing more and is free--on condition of respecting the customary contractual clauses Lasker (1950: 159 and 148). 3 This is a factor that is well translated in the dated but excellent report by Lasker (1950) by its significant title Human Bondage in South-East Asia. 4 A definition of slavery is usually summarised by the existence of two characteristics: 1) exclusion, 2) the fact that the master may draw a profit from him (see below). But this is not totally accurate: my own definition is that the slave is a dependant who possesses these two characteristics.
4 (advance notice of quitting, etc.)--to quit his job and his employer. As a worker, he is free. This is the situation that defines what we call free labour in the Western world. To clarify, the free labourer is not exempt from obligation, but remains obliged to perform the work for which he was contracted. But once the work is performed, once this obligation is acquitted, he is quit. By an act emanating from him and him alone, by labouring as stipulated in the hiring contract, he has left his situation, a situation that can be described as being under obligation (in French law, one who has an obligation to do, to give, or not to do). This is the contrast that was perceived and expressed by the scholars and administrators of Indochina we are examining. They certainly saw that all these unfortunate people of whom they were speaking, however hard they tried, could not leave their situation. They also clearly saw the difference--each says so in his fashion--between the debt-bondsman (gagй pour dette, generally called "pawn" in English) and the slave. The former can escape from this situation by an act emanating from him alone, by paying off his debt, which suffices to free him from it (once acquitted of his obligations, he is quit). And they were correct to say that someone who is tied by a debt is not a dependant, thus not a slave, for he can on his own account get out of this bind and does not depend on the master. He depends so little on him, moreover, that he can have the debt reimbursed by a Third Party
, whether his current master wants this or not,5 and so he will find himself tied to this third party but freed from his former master. It is often said that the contrast between the debt-bondsman and the slave is that the former represents only a temporary dependency or slavery, whereas the latter is dependent permanently. This is a crude approximation, because the true slave--even under the harshest form that can be conceived--can always be freed. The difference is that the debt-bondsman is free from his commitment--and from his obligation to work for his creditor--by the reimbursement of the debt, even if the master does not want it. Whereas the slave cannot be freed, except under the illusion that he "buys back" his freedom, but only if the master lets him. This means that even if both 5 Here is what Baudrit says (1942: 151), citing Postel: "...if he is discontented with his matser, he is free to borrow to pay him off and thus pass under a new domination..."
5 debt-bondsman and slave must similarly pay to be liberated, only the latter is a dependant (on his master), while the former is not (obtaining his liberty depends only on himself). This contrast appears quite clear, but it is only so under rather simple conditions, which we should now specify. Our writers--Baudrit, Silvestre, Briffaut, and the anonymous author--directly relate it to what were called the "sources" of slavery: on the one hand, insolvent debts that engender a situation that is not "truly" slavery but rather a form of debt-bondage (pawning); and on the other hand, various causes (criminal conviction, raids, etc.) that result in "true" slavery. But offered in such a radical and succinct form, this opposition is false, although our writers clung to it.6 But there is no reason why there should be an unequivocal correspondence between the source and the status of slavery, no reason why an insolvent debt corresponds solely to indebtedness, for in many societies debts can result either in mere indebtedness or in fact to veritable slavery. It is always a mistake to judge statuses from sources. To understand this we have to consider what is customarily called a slave. 1. CONCEPTUALISING THE MODES OF ENSLAVEMENT In most nineteenth century black African societies
there existed significant slavery that was much studied by Africanist ethnologists and historians, who taught us that the main characteristic of what Africans took to be slaves was that they existed outside any kinship, from which it followed that they had neither father nor mother (the master took the place of the father), that they could not be avenged by their family (which could decree that the person put into slavery--for debt, for example-- no longer belonged to it), that they did not have a place in the cemetery of the people of their tribe, etc. They were excluded from kinship. It is also clear that the principal feature of the ancient slave is that he was excluded from citizenship, and therefore he was not entitled to any rights. Slaves in the 6 Typical in this respect is a passage from Silvestre (1880: 275) that begins, "It is important for us to examine what the legislator means by slaves," which ought to introduce a discussion of the status (rights and duties) of the slave in Vietnam, but nothing of the sort follows, but instead an exposй of the causes (criminal conviction, sale, etc.)
6 Muslim world --as in medieval and modern Christianity-- were always pagans or infidels as of their capture, since both religions considered it illegitimate to reduce a believer to slavery. Pagans were excluded from the community of belief as soon as they were captured. Generalising from these facts, I have suggested the following definition of slave: A dependant: 1) whose legal status is marked by exclusion from at least one dimension that society considers to be fundamental; 2) and from whom profit may be drawn, in one way or another. Two notes are required. First, the legal status of a slave is very different from one society to another, but such a status always exists in any society. It is not the content--which is variable--of this status that is pertinent to this definition, but the existence of such a status. One should not confuse the fact of slavery--the fact of being a slave, that is to say, the fact of being held to be slave in a particular society--with its ordinary occupations or even its material living conditions
. There is almost no relation between the two. The material conditions of workers in the middle of the nineteenth century, with Women and Children working days
of 12 or 15 hours, were no better than those of ancient slaves working on latifundia, but nobody ever said that these workers were slaves. On the contrary, and with reason, one spoke of free labour. As for slaves, they did not always live in deplorable conditions, and there is nothing in common (at the material level) between a slave working in the galleys and the favourite slave of the master. Nor there is nothing in common with those slaves in Ancient Rome
who held important administrative posts or were famous actors who aroused fascination similar to today's pop singers. In truth, they did have one thing in common: the famous actor, the business manager of a rich master, and a concubine were all bound by the same laws that insisted that if they were called to testify in a trial, they could only do so under torture; if they had children, they could not give them their names, etc. They had in common only a legal status. Secondly, the basic social dimension from which they are excluded is equally variable according to the type of society. Exclusion from kinship in Africa, exclusion from citizenship in Rome --and in every kingdom they are excluded from an fundamental political relation because they
7 are not subjects of the king. As such, they do not owe taxes or military service. This was a very sure criterion of slavery, on which many scholars--Egyptologists, Orientalists or Hellenists--have insisted before me. For Vietnam: "Slaves are exempt from military service, and if they have no rights (in a share of communal land
, for example), neither do they have duties toward the state" (Nguyen Tung, 1998: 531). For the principalities of Laos: "No longer being free, the slave no longer pays taxes and cannot be recruited for the army" (Dorй 1998: 493). For Siam: "Now, a slave [a non-redeemable "that"7 the only type of that which I, following Lingat, take to be a true slave, see below] owes service only to his master. He is struck off the population rolls for the purpose of conscripted labor (corvйe). He is definitively lost for the service of the king" (Lingat, 1931: 83-86). Nunbhakdi adds that, "the royal decree of 1789 stipulated that only those that [in reality solely the non-redeemable ones] are exempt from corvйes" (1998: 472; it is traditional--although incorrect--to speak of corvйes in Thailand for taxes; it is probably this abuse of the medieval term specific to the West that leads this writer to an absurdity when he translates "phrai luang" as "royal serfs," while "phrai" refers in Thai to a free and thus taxable man).8 That a slave was not taxable was well-known in classical Antiquity, but is more rarely mentioned by Africanists. Taxation is not only a criterion of slavery, it is also a cause. The voluntary reduction into slavery for fiscal reasons appears as one of the key sources of slavery, particularly in Southeast and East Asia. We recall the Thai data. The subject owed six months of the year to the service of the king. The debt-bondsman owed only a third of his time, with another third owed to the lender. The slave owed nothing at all. We understand why the king closely regulated the contracts by which his subjects could sell themselves, their children or their wives into slavery. We also understand why subjects 7 That is a Siamese word and not the English demonstrative. 8 The data on ancient slavery among the Khmers, in the Angkorian or post-Angkorian period, are unreliable, as is the commentary by Marie A. Martin, who concludes that slaves were not tax-exempt (1998: 306), although one of the texts she cites states that "only" slaves attached to a religious foundation could be compelled to perform royal service, by which I understand that with this one exception (slaves of temples or monasteries always being a special case), all others could not be compelled.
8 could find some advantage in voluntarily making themselves slaves (Rabibhadana, 1969: 30 sq., 8889; Nunbhakdi, 1998: 465). The same motivations and same effects prevailed in Burma: "Slavery for debt attracted many people [...] in periods when the Crown had become too demanding: slaves were in effect exempt from taxes and services" (Brac de la Perriиre, 1998: 505). Fiscal data are uncertain for China after the Han period, especially for the dark times of the Three Kingdoms (220-280 A.D.). But the whole Chinese tradition shows peasants crushed by taxation and liable for military service of excessively long duration. So we should not be surprised to find a Sinologist like Balazs (Maspero and Balazs, 1967: 98-99) envisaging fiscal flight as one of the principal causes of the growth of slavery--or other forms of dependence--during the troubled period that followed the fall of the Han dynasty. There was sometimes a benefit to being a slave. 2. PLEDGING HUMANS AS COLLATERAL FOR DEBT IN SIAM The definition I have given of slaves allows us to differentiate them from other dependents such as serfs in the Middle Ages, or helots from Sparta or Ancient Greece. It also allows us to differentiate a slave from a pawn, which I will now explain. Africanists have long pointed to a phenomenon called "pawning," which consists of placing someone with a creditor as collateral for a debt or as security for a loan. The "pawn," sometimes called a "hostage" or more rarely a "pledge," is at the service of the creditor and owes him all--or almost--of his time. This form of servitude has often been confused with slavery for debt, all the more easily in that the pawn risked eventually being reduced to slavery--and in practice was, more often than not--if the debt was not reimbursed. Yet these are totally different institutions. In effect, the pawn does not possess the dubious quality that is one of the decisive criteria of the slave: he is not excluded from his kinship, he still belongs to his lineage, he keeps his name, he can participate in the lineage council and in the management of lineage business, he participates in their rituals, he can marry and have legitimate children. The person with whom the pawn is placed and who has so many rights over him or her--right to labour, often right to have sexual relations when it is a woman--does not have, unlike the usual situation of master to slave, the right of life or death over his dependent, and only a limited right to punish. Finally, everyone
9 placed as a pawn would be immediately freed upon payment of the debt. This represents another difference from the situation of a slave, who can be redeemed only if the master consents. By contrast, the pawn is freed by the reimbursement of the debt, even if the person with whom he was placed does not consent. And yet pawning a person represents a form of enslavement that seems particularly harsh. The fundamental principle of pawning means in effect that the labour, services and tasks of all kinds that the pawn supplies do not contribute to the reimbursement of the debt because of which he or she was pawned in the first place. The debt, in other words, is not eliminated or lessened by the pawn's labour. Actually, it often happens that it increases due to the interest that continues to accumulate if it is not reimbursed, and without any labour reducing this interest. The consequence is obvious: the pawn cannot generally free himself, and has to work his whole life for a debt that may have originally been quite slight. The complexity of pawnship for debt therefore arises from the paradox that the pawn remains by law a free person,9 meaning he remains member of his kin, with all the consequences this belonging implies, enjoying rights, etc. By law, he is still capable of freeing himself by paying back the amount of the debt. However, the pawn is in fact enslaved, often with no hope of ever being able to free himself, and living in material and social conditions analogous to (or even worse than) those of a slave.10 With the difference between a slave and a pawn now clarified, we should now examine why it is more murky in practice. First, situations are likely to evolve, with the pawn being liable to change status and fall into the category of slave. Thus we have the same source--debts--as the possible origin of two different states. This is what happened in ancient Siam, certainly the best-known 9 In particular there is no status in the legal sense of a pawn, unlike for a slave. His general status remains that of a free person, but bound by obligations. While the slave belongs in the framework of what we are calling legal (or statutory) dependency, the pawn belongs in the framework of de facto dependency. 10 This comes from the fact that the creditor-pawnbroker cannot adopt the pawn, a common fate of slaves in lineage societies or even in Imperial Rome.
10 example in Southeast Asia,11 thanks in particular to the massive study by Robert Lingat (1931) of the 1805 laws, written in Bangkok just after the destruction of Ayutthaya and its archives in the Burmese war, but which used information from the Ayutthaya period (1361-1767). Thai law recognized the existence of a category of person, the that, which Western commentators render as "slave" but which, for reasons explained below, I would prefer to translate by the deliberately vague term "dependant." This same law distinguishes classically, at least among the "bought" thats, between redeemable and non-redeemable thats. The former are those who could buy themselves back from their master even against their will, since they possessed this right, and the amount they paid was the same as the master paid to acquire them. A written document--the kromathan--guaranteed both the that's right and the master's temporary possession against possible flight. It is clear, as Lingat explains, that the sale that made them thats was a fiduciary sale, i.e., a repurchase agreement. They were in exactly the same position as the pawned person:12 their labour is due to the master until the sum for which they are sold is reimbursed, but if they manage to gather it by themselves (or any relation, friend, or whomever is well-disposed to him), then the redeemable that could be disengaged from the hold of his master. The 11 Among the first Western observations is the book by La Loubиre, Du Royaume de Siam, (Paris: J.-B. Coignard, 1691), written when Louis XIV
sent an embassy to the court of Ayutthaya (1687-1688), which is the most famous description. For the nineteenth century, see Bowring (1857 vol. I: 189-199), reprinted by Lasker (1950: 283-288). More recently, after Lingat's study, Rabibhadana (1969: 104-112); Turton (1980); Terwiel (1983). 12 For a systematic comparison of the redeemable that and the pawn, see Lingat (1931: 51 sq.). He does note one difference: the situation of the former is more favourable than the latter because the pawn normally falls into the creditor's patrimony as of date of payment, whereas the right to buy oneself back seems to be inalienable for the redeemable that. We note that the two factors in placement as pawn (labour does not reimburse the debt, but the pawn can always liberate himself by furnishing the amount of the debt) are perfectly in evidence in these lines Alfred Raquez (cited by Dorй 1998: 490) devoted to those pawned for debts in Laos at the end of the nineteenth century: "The debt cannot be eliminated totally
11 labour of those placed in bondage represented interest, for their labour did not accrue to reduce what they must have paid to buy themselves back, any more than this sum could increase over time. In fact, it is clear that the redeemable that is not a slave at all: not only does he have the right to buy himself from the master, but also he remains--unlike the non redeemable that--a subject of the king and as such pays taxes. His condition is specially protected, particularly by strict limits on the master's right to punish him. He is part of the master's patrimony, but only for the amount of the sum by which he was acquired.13 This fiduciary sale with right of repurchase is opposed to a firm sale, "definitive" or "pure and simple," as Lingat says. In this case, there is no kromathan and the non-redeemable that can buy himself or be buy back, but only if the master agrees (whereas the redeemable possesses this right over his master). In addition, the master's right to punish is much more extensive. Finally, he is no longer a subject of the king and owes no taxes or military service. These non-redeemable thats can be seen as slaves in the full sense of the term, becoming such by being buy and sell. Any free person is free to sell himself, or any child or spouse, both of whom lie under the power of the father or husband. Reading the laws of 1805 leaves no doubt on this matter: selling into slavery was in Thailand a common phenomenon,14 normal and perfectly legal until its partial abolition in 1873. Although the law on loans and details of the ancient procedure are not totally known, the insolvent debtor could be seized and sold, and apparently entered into the category of non-redeemable that, at least in the Bangkok Period (1763-1932).15 But it is unlikely that the more ancient law was any more favourable except by chance, for the labour of the slave is considered as compensation for his nourishment and is not deducted from the sum due [...] Slaves for debts are liberated by the payment of their debt [...]." 13 This last point is well documented by Lingat (1931: 164) who also says that fiduciary sale did not put the sold that under the power (siddhi, a term that is the analogue of the Roman potestas) of the buyer. 14 Five articles are devoted (sections 42 to 46, pp. 323-326 of Lingat's translation) to this "definitive" sale as opposed to the fiduciary sale, which leaves no doubt as to its legality. Apparently, it fetched a sum higher than the agreed price for fiduciary sale. 15 Lingat (1931: 46). There is the same uncertainty about the reducible nature (or not) of someone who is rescued in times of dearth and thereby becomes a that in conformity with the Indian tradition (ibid., 44).
12 to the indebted, as the general evolution of Thai legislation tended toward the softening of the slaves' condition, and finally to the abolition of the institution. So everything leads us to think that slavery (in the full sense of the term) for debts existed in Siam, and that slavery was completely legal. Apparently the same was true in Burma and in Cambodia, and perhaps in Laos, that is, in the whole region marked by the influence of India, which in some of its laws perfectly accepted the legitimacy of "slavery from hunger," just as it accepted the legitimacy of reducing people to slavery for the cause of debts, at least among the lowest castes. But things were different in areas marked by the influence of China: Vietnamese or Chinese legislation did not accept the legitimacy of selling children into slavery,16 nor selling oneself into slavery, nor slavery for debt. Baudrit, Silvestre, Briffaut and the anonymous author clearly saw this. The first makes this particular remark (Baudrit, 1942: 151): "Philastre [translator of the Annamite code, speaking of Cochinchina], who was writing in 1865, said these ideas of slavery being freely consented to [by the pauper or the debtor] were in contradiction with the spirit of the [Annamite] law; they were no less so in Cambodia, where a royal decree of 1877 merely softened the rigours." 3. PLEDGING LABOR AS COLLATERAL FOR DEBT IN VIETNAM Vietnamese laws did not admit slavery for debts, or even the phenomenon of the pawn. This is something that should be stressed, and there is no better way than to summarise the study by Dang Trinh Ky, L'Engagement des Personnes en Droit Annamite [The Pawning of Persons in Annamite Law] (1933). This notable example will allow us to show the difference, in both the spirit and principles of a society and its law, between the institutional possibility of paying back a debt by one's labour (which is the case in ancient Vietnam) and the pawning of a person (in which the principle is that the labour does not liquidate the debt). The author refers essentially to the Le dynasty (fifteenth to eighteenth century), and occasionally to the Gia Long code promulgated in 1812.
13 1) The labour performed by the debtor for the creditor decreases the debt at a rate fixed by law: "for a loaned sum from ten to 20 ligatures (each ligature worth 600 sapиques),17 the reimbursement is 17 sapиques per day; for 21 ligatures on upward, it will be 23 sapиques;..." and so on, with the rate of legally fixed amortization increasing with the amount of the debt, a measure that is favourable to the debtor (Dang Trinh Ky, 1933: 115).18 2) As Dang Trinh Ky clearly explains (ibid., 54), this institution is for "hiring services," meaning a form of waged labour in which the amortization of the debt counts as salary. I stress that this is the key point of the whole institution: Annamite law recognizes the value of labour, whereas pawning does not. 3) It is not the person that is pawned. The resemblance with pawning remains superficial: if the debtor resides with the creditor while he labours for him, this is because in traditional Annam territory any labourer resides with his employer. However, the creditor does not have rights over this person. 4) Dang Trinh Ky (ibid., 54) insists on the fact that Annamite vocabulary never refers to the engagement of the debtor by the term "sale" or even "fiduciary sale" or "repurchase agreement," but rather as "rental." 5) On this point, there is an essential difference in Annamite law between persons and things: only the latter may be "sold," the former may only be "rented." 6) Nor can the creditor sell or rent out the debtor who is placed in his service: "In effect, he cannot sell the engaged person on the model of the Roman or Siamese creditor. [...] The lender here does not have ownership over the pawned person. Not only does he not have 16 "Someone who sells his children or grandchildren as slaves is punished with 80 strokes of truong" (Silvestre, 1880: 226). A truong was a medium size wooden stick. 17 The sapиque is the fomer Chinese or Indochinese currency. 18 The rate of interest is fixed at 15 sapиques per ligature per month, which although significant always remains inferior to the value of the labour that the debtor may furnish. Moreover, the law envisages a "bonus for labour."
14 ownership, but he does not even have paternal power over the individual,19 he cannot rent him or use him as security." (ibid., 85) 7) The debtor's engagement with the creditor cannot be an engagement for life. First, the law seems to require that the term be fixed. The contracting parties usually stipulate a maximum period, generally three to five years. In any case, the laws on amortization of debt mean that it will one day be automatically repaid. In Annamite law there is no life-long engagement.20 8) Finally, in the course of his study Dang Trinh Ky occasionally recalls some major principles of Annamite law, of which two are worth citing. First, slavery has only three primary source
s: war, judicial conviction, and birth. Slavery for debt does not exist. Second, the sale of children by parents is rejected as an abomination, as is sale of wives by husbands. These points demonstrate that the treatment of debt conforms to the general character of law: never can a person of free status be transformed into a permanent dependent solely for financial reasons. Thus there is a very clear opposition between two cultural traditions
, two systems of law: Indian and Chinese. But of course ordinary people ignore legal subtilities, people who in ancient Chinese tradition "do not know the rites," meaning do not know the proper customs. They disregard the edicts of a distant government, often incapable of sanctioning its decrees other than by a few exemplary but rare punishments. The people of China, especially southern China, as well as of Tonkin, Annam, and Cochinchina had since time immemorial sold their daughters to whoever wanted them, who made of them whatever they wanted, whether becoming concubines or "being put on the river as a prostitute" (Jachok 1988: 146-7); occasionally they also sold their sons. When there was no other solution, the head of the family would abandon himself into the hands of another for as long 19 An allusion by the author to sons placed in mancipium in ancient Roman law, a placement that transferred the father's power (potestas) to the buyer. 20 Here again the resemblance to the modern notion of the contract appears: contracting parties can only be engaged for a limited duration.
15 as he could be paid and thus spport them. These peoples would not behave differently than those of Thailand, Cambodia, Burma or India. Everywhere poverty led to slavery. 4. LEGAL PLURALISM: THE MUI-TSAI CASE IN SOUTHERN CHINA This is what Baudrit, Silvestre, Briffaut, and the anonymous author were passionately discussing, as moralisers but also as good observers of social life. They knew that in the Far East things were happening that had no equivalent in all of Western history. For example, a Shanghai newspaper of 1930 reported that at this time famine was so severe that around Lishan, in Southern China, thousands of children were brought by their families in the hopes of selling them rather than dying of hunger (Lamson (1934: 562-563). What Baudrit, Silvestre, Briffaut and the anonymous author were testifying to was this permanence over time, beyond legislation and forms of government, of this consistency of men and women to sell themselves, despite what we call "freedom," in order to live or help their families live. The haunting and basic question remains: whether this is indeed slavery, or whether this word was used to move good souls in the West by informing them about the miseries in the Far East? In my mind there is only one way to answer this question intelligently, and this means to mobilise a strong notion recently in favour in legal circles: legal pluralism. A state has its laws, which define the legal level, but people have their own norms that are not necessarily the same, to which they cling and which they respect by force of habit, but also because they hold them as legitimate. Criminal mafias and interlopers of all kinds also have their norms, which have nothing to do with those of the state. The example of the mooi-jai (mui-tsai) can serve as illustration.21 In Cantonese dialects, the term signifies "little sister" or "little servant" and was allied to all those who had been sold when very young by their parents. Their number was estimated at two million in the 1920s for all of China. International opinion was moved and launched a campaign, called the anti-mui-tsai campaign, to stop 21 Kulp (1925: 165); Lamson (1934: 562-566); Lang (1946: 259 ff); Watson (1980 and 1991: 245); Hershatte (1991: 266); Jaschok 1988. For the sale and resale of wives, see the Institute of Pacific Relations (1938: 84).
16 this human traffic and to free what were now called "little slaves." In reality the sale of humans in China--as in many non-Western societies--included various phenomena. The very appellation mooijai corresponded to very different realities,22 which ran from the "concubine" or secondary wife bought for a sum of money (analogous to what is described in anthropology as bride price or bridewealth), to the impoverished girl placed to labour as servant to a richer family, to the one placed as a security for a loan, etc. So it was wrong in the West to assimilate indiscriminately all mooi-jai to slaves. This vast ensemble included a varied palette of contrasting situations from the standpoint of both status and living conditions. But it is undeniable that within this ensemble, some women could be described as being in slavery. Witness this contract of 1927, authenticated by affixed fingerprints, concerning a girl named Ah Mui: In consequence of urgent needs for funds to meet family expenses, I am willing to sell my own daughter, Ah Mui, ten years of age [...] to Chan Yee Koo through a go-between. [...] Price is to be 141 dollars. After this sale, Chan Yee Koo shall have the right to change the name of the girl. If the girl is disobedient, Chan Yee Koo shall be allowed to resell her, and the mother shall have no recourse. In the event of any misfortune befalling the girl, there is no blame to either Party. It is also perfectly clear that the girl has neither been betrothed to any other family, nor is there any mortgage on her. [...] This is a straightforward sale and purchase.23 Another, much longer contract dated 1895 has the same details: fingerprints, presence of a gobetween, certification that the girl is neither betrothed nor mortgaged, firm sale and not for repurchase, price demanded, non-responsibility of both parties in case of misfortune befalling the girl, etc. The terms of this sale stipulate: 22 Well attested by Jaschok (1988: 8 ff). 23 Ibid., p. 146-147.
17 The makers of this deed for the absolute sale of a girl for the purpose of prostitution were the Tang family, who had a girl for prostitution surnamed... and named... aged 16. [...] They wished to sell her to anyone, no matter whether living near them or at distance, or whether living on land or afloat. The price demanded was 270 dollars. Through go-between Li Shi, the girl was taken to... who examined her and found her all right, and agreed to pay the amount demanded, viz. 270 dollars, at the rate of 71 to the dollar. Matters were explained in the presence of the three parties, and the two parties mutually agreed to the bargain. A deed was drawn up for the transaction that day in the presence of the go-between, and both the girl and the deed and the full price were handed over and all matters concluded. This girl has never been betrothed to any family. When the sale is complete, the girl is to go away forever. [...] The purchaser has a right to have the girl taught to play music and to sing in order to that she may be put on the river as a prostitute, and she will dress herself up to receive visitors as her calling and thus spend her life [...].24 These ancestral practices, well known to historians, were illegal.25 Chinese authorities had outlawed them long ago. The Qing Code (Ts'ing, 1644-1911) devoted several articles to the repression of the pawning or renting of a daughter or a wife, and to the sale "for principal or secondary wife or slave" of a daughter, sister or wife Boulais (1924: 265 ff). But there would have been no need to do so if this had not been a common practice. We can even go farther in relation to what was just said about legal pluralism. This illegality (illegitimacy from the standpoint of the law) undoubtedly coexisted with the contrary idea that it was legitimate for parents to sell their children. This idea or sentiment was rooted in the fact that this was an old practice anchored in custom, and ultimately in a morality (just as ancient) that gave the father omnipotence within the family. Secondly, 24 Ibid., p. 147. 25 Gernet (1959: 160); Wang Yi-T'ung (1953: 313-314).
18 the interdiction by the state was on a par with the Confucian morality that did limit the power of the father (since antiquity it had been forbidden in China for a father to kill his son) and that strongly reproved making a profit on one's kin (for example, receiving money for the marriage of a daughter was "compensated" by granting a dowry equal or higher in value). But this morality was that of the elite and not the peasants, who accepted "making money" by marrying their daughters, placing them as servants, or even selling them for disreputable purposes. But what is most striking about these examples is that, totally illegal as they may be, these contracts assumed a legal form. These are contracts, in proper and due form. There is indeed an ordinary or popular morality that is not that of the state but is still a morality, almost a right, with its norms, procedures, etc. And this quasi-right, this extra-legal legality, is related to the old conception of the slave as excluded from his original relations, which the contracts specify: the new master and owner of the girl will have the complete right to change her name and ultimately her identity. Is this what we could call de facto slavery or rather slavery according to its status in inferior but popular law? These mui-tsai contracts began to become known, in the academic world at least, thanks to the work of Maria Jaschok and others. It is interesting to find completely similar deeds of sale coming from Cochinchina in Silvestre's report (1880: 150): We the undersigned [...], living in the village of An Thanh, canton of Bao An, district of Ben Tre, declare that, having fallen into extreme poverty and not being able to pay the tax we owe to the state, are definitively selling our son Ngoc, aged 16 years, to he who is named Ven, of the village of Bhia An Trung, for the sum of 20 piastres. Our son will henceforth be part of the buyer's family and will bear the family name of the adoptive father. If later on difficulties occur, we must bear the consequences. However, we reserve the right to repurchase our son for the sum that we are receiving, plus the interest. Made the first day of the fourth month of the year Ky Mao (1879).
19 We have the same formalism, same contract written and notarised (which is significant for probably illiterate peasants), the same mention that the buyer can have the purchased son bear his name. These illegal usages sit strangely alongside other legislation where they are legal. But there is one slight contradiction in terms, for one cannot sell one's son "definitively" and at the same time reserve the right to buy him back. Let me summarise and conclude. The complexity of slavery in Southeast Asia appears formidable and it defies any simplistic attempt to classify forms of dependence. This is not due only to the multiplicity of legal statuses between freedom and slavery including the pawn, of which our legal system is practically ignorant. Nor is it only because the constraints of the situation (an insoluble debt) might make a person legally pass from one status to another. Nor is it only because there exist practices that are well anchored in tradition and therefore beyond the law -- if not perfectly illegal and combated as such by central powers. It is because this tradition seems to have shaped these illegal forms into norms that copy or mimic legal forms, so well that, underlying the official list of legal statutes, we must perceive others, sorts of doubles halfway between the law and the facts. Or should we invoke the exceptional creativity of Oriental social life with respect to forms of dependence? We have to find other names for other kinds of slavery. Translated by Susan Emanuel References Cited Anonymous, Revue Maritime et Coloniale, no. vii (1863): 78-80. Baudrit, Andrй, Bйtail Humain. Rapt-Vente-Infanticide Dans l'Indochine Franзaise et Dans la Chine du Sud. (Saigon: Sociйtй des Imprimeries et des Librairies Indochinoises, second ed., 1942). Boulais, Guy, Manuel du Code Chinois. (Chang-Hai: Imprimerie de la Mission Catholique, 1924).
20 Bowring, John, The Kingdom and People of Siam. (London: Kessinger Publishing, 1857, reprinted in 2009). Brac de la Perriиre, Bйnйdicte, "Note sur l'Esclavage en Birmanie d'Aprиs le Livre de Manu", in Formes Extrкmes de Dйpendance: Contributions а l'Etude de l'Esclavage en Asie du Sud-Est, ed. Georges Condominas. (Paris: Editions de l'EHESS, 1998), pp. 499-507. Briffaut, Camille, "L'Engamement pour Dettes Dans la Droit Annamite", extracted from Questions Pratiques de Lйgislation et d'Economie Sociale. (Lyon: Imprimerie A. Storck et Cie, 1907) Dang Trinh Ky, L'Engagement des Personnes en Droit Annamite. (Paris: Domat-Montchrestien, 1933). Dartiguenave, Henri, "Des Ventes d'Enfants en Indo-Chine", Revue Indochinoise, no. 76, (1908): 239-247 Dorй, Amphay, "Aperзu sur l'Esclavage au Laos (XIII-XIX siиcles)", in Formes Extrкmes de Dйpendance: Contributions а l'Etude de l'Esclavage en Asie du Sud-Est, ed. Georges Condominas. (Paris: Editions de l'EHESS, 1998), pp. 483-498. Gernet, Jacques, La Vie Quotidienne en Chine. (Paris: Hachette, 1959). Hershatte, Gail, "Prostitution and the Market in Women in the Early Twentieth-century Shangai", in Marriage and Inequality in Chinese Society, eds. Rubie S. Watson & Patricia Buckley Ebrey. (Berkeley: University of California Press, 1991), pp. 256-283. Jaschok, Maria, Concubines and Bondservants: The Social History of a Chinese Custom. (Hong Kong
& London: Oxford University Press & Zed Books Ltd, 1988). Institute of Pacific Relations, Agrarian China: Selected Source Materials from Chinese Authors (Chicago: University of Chicago Press, 1938) Kulp, Daniel Harrison, Country Life in South China. The Sociology of Familism. (New York: Bureau of Publications & Teachers College
, Columbia University
, 1925). Lamson, Herbert Day, Social Pathology in China. (Shanghai: Commercial Press, 1934).
21 Landes, Antony. "Lettre au Gouverneur. Rapport sur la Prostitution", Excursions et Reconnaissances vol. 2, no. 4 (1880): 145-147 (annex of Silvestre, Jules, Excursions et Reconnaisances vol. 2, no. 4 (1880): 95-144). Lang, Olga, Chinese Family and Society. (New Haven: Yale University Press, 1946). Lasker, Bruno, Human Bondage in South-East Asia. (Chapel Hill: University of North Carolina
Press, 1950). Lingat, Robert, L'Esclavage Privй dans le Vieux Droit Siamois. (Paris: Domat-Montchrestien, 1931). Martin, Marie-Alexandrine, "L'Esclavage au Cambodge aux Temps Anciens d'aprиs les Sources Epigraphiques et Bibliographiques", in Formes Extrкmes de Dйpendance: Contributions а l'Etude de l'Esclavage en Asie du Sud-Est, ed. Georges Condominas. (Paris: Editions de l'EHESS, 1998), pp. 285-314. Maspero, Henri & Balazs, Etienne, Histoire et Institutions de la Chine Ancienne des Origines au XIIe Siиcle Aprиs J.-C. (Paris: Presses Universitaires de France, 1967). Nguyen Tung 1998 "L'Esclavage dans le Viкtnam Ancien", in Formes Extrкmes de Dйpendance: Contributions а l'Etude de l'Esclavage en Asie du Sud-Est, ed. Georges Condominas. (Paris: Editions de l'EHESS, 1998), pp. 509-540. Nunbhakdi, Sutwahadee, 1998 "Etude sur le Systиme de Sakdina en Thaпlande", in Formes Extrкmes de Dйpendance: Contributions а l'Etude de l'Esclavage en Asie du Sud-Est, ed. Georges Condominas. (Paris: Editions de l'EHESS, 1998), pp. 459-481. Rabibhadana, Akin, The Organization of Thai Society in the Early Bangkok Period 1782-1873. (Ithaca: Cornell University
, Southeast Asia Program Data, Paper n. 74, 1969). Terwiel, Barend, "Bondage and Slavery in Early Nineteenth Century Siam", in Slavery, Bondage and Dependancy in Southeast Asia, ed. Anthony Reid. (Santa Lucia: University of Queensland Press, 1983), pp. 118-137. Silvestre, Jules, Excursions et Reconnaisances vol 2, no. 4 (1880): 95-144. Testart, Alain, "L'Esclavage pour Dettes en Asie Orientale", Mousson 2 (2000): 3-29.
22 Testart, Alain, L'Esclave, la Dette et le Pouvoir: Etudes de Sociologie Comparative. (Paris: Errance, 2001). Testart, Alain, "The Importance and Significance of Debt Slavery", Revue Franзaise de Sociologie, Supplйment (2002): 173-204. Turton, Andrew, 1980 "Thai institutions of Slavery", in Asian and African Systems of Slavery, ed. James L. Watson. (Oxford: Blackwell, 1980). [Reprinted in Formes Extrкmes de Dйpendance: Contributions а l'Etude de l'Esclavage en Asie du Sud-Est, ed. Georges Condominas. (Paris: Editions de l'EHESS, 1998), pp. 411- 457]. Wang Yi-T'ung, "Slaves and other Comparable Social Groups During the Northern Dynasties (386618)", Harvard Journal of Asiatic Studies 16 (1953): 293-364. Watson, James L., "Transactions in People: The Chinese Market in Slaves, Servants, and Heirs", in Asian and African Systems of Slavery, ed. James L. Watson. (Oxford: Blackwell 1980), pp. 223-250. Watson, Rubie S., "Wives, Concubines, and Maids: Servitude and Kinship in the Hong Kong Region, 1900-1940", in Marriage and Inequality in Chinese Society, eds. Rubie S. Watson & Patricia Buckley Ebrey. (Berkeley: University of California Press, 1991), pp. 231-256.