In the afterglow of the 2008 bicentennial commemoration of the British abolition of the slave trade in its colonies, the United Kingdom has been reminded repeatedly that slavery persists to this day. As then Prime Minister Tony Blair noted in the lead up to the bicentenary, ‘we also need, while reflecting on the past, to acknowledge the unspeakable cruelty that persists in the form of modern day slavery’.Footnote 1 Since then we have seen the introduction of provisions dealing with slavery, servitude and forced or compulsory labour in the Coroners and Justice Act 2009; and in 2011, the Home Secretary launched Human Trafficking: The Government's Strategy, meant to address organized crime networks seeking to move people into the United Kingdom to exploit them by way of the crimes set out in the Coroners and Justice Act.
It might seem curious that, having left behind the slave trade in the age of sail, indeed, having led that abolitionist movement, and having determined in 1772 it was ‘a soil whose air is deemed too pure for slaves to breathe in it’, the United Kingdom should now find itself for the first time having actually to legislate against slavery.Footnote 2 While this is curious indeed, more pressing from a legal perspective is that we do not have a clear understanding in law of what the term ‘slavery’ actually means. It is generally accepted that the legal definition of slavery is provided by Article 1(1) of the 1926 League of Nations Slavery Convention, which reads: ‘slavery is the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised’. However, as we show here, in recent times the term ‘slavery’ has been the subject of misinterpretation to the extent that it has been rendered so wide as to be effectively meaningless. Our aim is to reclaim the legal definition as one with a core settled meaning which has demonstrable relevance to contemporary practice. To make this argument, we use property law perspectives to explain the conception of ownership invoked by Article 1, and to articulate its continuing relevance to the modern world. We argue specifically that it entails a substantive as opposed to a formal understanding of slavery, such that slavery will exist today in any case where the provisions of Article 1 can be satisfied on the facts.
I. THE PROBLEM OF INTERPRETING THE DEFINITION OF SLAVERY
In law, the baseline for understanding slavery is the definition set out in Article 1(1) of the 1926 Slavery Convention. The Vienna Convention on the Law of Treaties notes that in seeking to interpret a provision, account should be taken, inter alia, of ‘any subsequent agreement between the parties regarding the interpretation of the treaty or the application of its provisions’.Footnote 3 In the case of Article 1 of the 1926 Convention, subsequent agreements confirm its status as the authoritative source on what counts as slavery in law.Footnote 4 Thus, prior to the 1956 Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery, the question of the adequacy of the 1926 definition was raised by the Ad Hoc Committee on Slavery, a United Nations expert body established to consider the overall issue of slavery. Its determination was that there was ‘not sufficient reason for discarding or amending the definition of slavery contained in Article 1 of the International Slavery Convention of 1926’. The Committee therefore recommended that ‘this definition should continue to be accepted as an accurate and adequate definition of the term’.Footnote 5 That recommendation was carried throughout the process which ultimately saw the inclusion of the 1926 definition of slavery as a provision of the 1956 Supplementary Convention.Footnote 6 Likewise, in the process which ultimately saw the establishment of the International Criminal Court in 2002, the issue of the definition of the crime against humanity of ‘enslavement’ was open for discussion. During the deliberations of the Preparatory Committee on the Establishment of an International Criminal Court in 1996, it was noted that some ‘delegations expressed the view that enslavement required further clarification based on the relevant legal instruments’.Footnote 7 Although never explicitly stated in the negotiations leading to the 1998 Statute of the International Criminal Court, States, having given themselves the option to rework the definition, decided to leave the substance of the 1926 definition of intact.Footnote 8 Thus, later agreements show that ‘the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised’, is established as the contemporary, legal definition of slavery.
If the locus of the legal definition of slavery can easily be settled, the question of how the provisions of Article 1 are to be interpreted has proved much more controversial. A general problem has been that the use of the term ‘slavery’ has often been appropriated to give visceral emphasis to incidents and issues which go far beyond the parameters of the definition or any plain reading of its text. Certainly this has happened in sensationalist newspaper reports, but it also figures in legal scholarship on slavery. For example, Jane Kim has argued that rape should be equated with slavery for legal purposes, since rape victims ‘at the time of their assault and perhaps forward are subject to the domination and degradation by [sic] another person’.Footnote 9 The use of Kim's piece is illustrative of a general interpretative malaise in defining slavery, and usefully points to others who have written about, as Kim puts it, ‘less institutive forms of violence as constituting slavery’, citing: child abuse, child soldiering, domestic violence, forced prostitution, mail-order brides, prohibited abortions, and sexual harassment in housing.Footnote 10 And less one be left with the impression that such equating of slavery is the exclusive domain of American law reviews, it should be emphasized that such comparisons are in good company. During much of 1990s, the United Nations’ Working Group on Contemporary Forms of Slavery (the de facto guardian of the 1926 and 1956 Conventions) considered the following issues as falling under its rather expansive mandate: child pornography, children in armed conflict (1990); child soldiers (1991); removal of organs (1992); incest (1993); migrant workers, sex tourism (1994); illegal adoption (1996); and, early marriages and detained juveniles (1997).Footnote 11
This desire to interpret so widely the legal definition of slavery stems from a failure of States to prosecute and try individuals for enslavement during much of the twentieth century. Into this vacuum came parties, both NGOs and Governments, which oftentimes had a vested interest in expanding the notion of slavery—thus diluting the legal definition—so as to benefit from the visceral call to end a particular practice. The result has been ambiguity in the legal definition, but also ambivalence in the value or usefulness of that definition. This dilemma is evident even in Suzanne Miers's leading text on issues of slavery during the twentieth century. On the one hand, in considering slavery, Miers notes that the UN ‘Working Group was using the term to cover such a wide range of practices as to be virtually meaningless’ and goes on to press the need for a new definition, declaring that only ‘a new clear definition in international law could untangle the morass, set clear standards, and allow governments to prosecute offenders, and victims to seek redress through the courts’.Footnote 12 On the other hand, for Miers, the requirement for a ‘new clear definition’ of slavery is predicated on the welcome fact that by ‘widening the definition of slavery to include “all its forms”, it marked the beginning of the attack on a wide range of exploitive practices, and was the first step in an ever-expanding definition, as more and more practices came to be described as “analogous” to slavery’.Footnote 13
Yet, if we return to the actual definition of slavery, the phrase ‘all its forms’ is not found. Instead of considering the actual definition, Miers follows a well-trotted precedent of interpretation which goes beyond the definition of slavery as set out in Article 1 of the 1926 Convention, and instead reads into the definition other elements of the 1926 Convention to allow for an ever-expanding notion of slavery. For the sake of clarity: it is not the 1926 definition of slavery which has caused an ever-widening concept of slavery to emerge, but a specific interpretation, by non-legal entities, that has gone beyond the definition as set out in Article 1 of the 1926 Slavery Convention. The genesis of this line of interpretation of the definition of slavery starts with the 1930 International Commission of Inquiry meant to look into whether or not slavery or forced labour existed in Liberia.Footnote 14 The International Commission felt constrained in its inquiries as its remit fell to ‘whether slavery as defined in the anti-slavery convention in fact exists in the Republic’. As such, beyond slavery itself, the Commission noted, the 1926 Convention did not allow for a consideration of lesser servitudes, as ‘in its present form it leaves little room for distinguishing between degrees of restrictive freedom’. Having failed to find slavery, the International Commission of Inquiry moved to justify a wider understanding of slavery so as to be able to investigate forced labour which was then persisting in Liberia. As a result the International Commission provided a novel interpretation of slavery by going beyond the definition and noting that there were ‘two important provisos’ found in the 1926 Slavery Convention: ‘to bring about … as soon as possible the complete abolition of slavery in all its forms’ and ‘to take all necessary measures to prevent compulsory and forced labour from developing into conditions analogous to slavery’.Footnote 15 With this in mind, the International Commission redrew its mandate, stating that ‘in the interest of achieving the objective of the convention “to bring about the abolition of slavery in all its forms”, it has decided to group its evidence which shall be the basis of its findings under the following divisions: A. Common slavery [… and] B. Oppressive practices restrictive of the freedom of persons, constituting conditions analogous to slavery and tending to acquire the status of common or classic slavery’.Footnote 16
We will see shortly that this interpretation of the 1926 Convention by the International Commission of Inquiry was dubious at best, resting in part on a misreading of the Report which accompanied the tabling of the Convention before the League of Nations in 1926;Footnote 17 but it certainly did gain credence as an accepted understanding of slavery throughout the twentieth century. This is most evident in the 2000 Working Paper of the now defunct UN Sub-Commission of Human Rights, written by David Weissbrodt and Anti-Slavery International, which turned to this notion of abolition of slavery in all its forms to anchor its understanding of the definition of slavery. The Working Paper argues that by ‘referring to “any or all of the powers of ownership” in its definition of slavery, and setting forth as its stated purpose the “abolition of slavery in all its forms” the Slavery Convention covered not only domestic slavery but also the other forms of slavery listed in the Report of the Temporary Slavery Commission’.Footnote 18 As such, for Weissbrodt, the 1926 definition of slavery includes so-called ‘other forms of slavery’ such as serfdom and debt bondage.Footnote 19 Such an interpretation originally found a footing in the United Nations from mid-1960s onwards, and gained momentum as a consequence of a concerted effort to expand the definition of slavery so as to include apartheid.Footnote 20 In this climate, an interpretation like that utilized by Weissbrodt met the political objective of the United Nations, but, as with the International Commission of Inquiry, fails to appreciate the legal definition of slavery.
On legal grounds this expansive interpretation of the definition of slavery simply does not hold. It is clear that States, in negotiating the content of the 1926 Convention, did not intend to widen the scope of slavery by reading into it other types of exploitation which failed to meet the threshold of the exercise of ‘any or all of the powers attaching to the right of ownership’.Footnote 21 In presenting the Convention to the Assembly of the League of Nations for its adoption, the Rapporteur noted that provisions which would have included lesser servitudes had been excluded as a result of comments received back from States, it being understood that if certain practices met the threshold of the definition of slavery, they came within the preview of the Convention, and yet ‘if these last practices do not come under the definition of slavery as is given in Article 1, the [Temporary Slavery] Commission is unanimously of the opinion that they must be combated’.Footnote 22 And combated they were, though only later, in 1956, by the United Nations which brought into being a Supplementary Convention to addresses specifically those servitudes mentioned by Weissbrodt, as developed by the Temporary Slavery Commission.
With the United Nations supporting a diluted notion of slavery and a failure of case law to emerge during the twentieth century, the dawn of the twenty-first century has seen renewed emphasis on the Article 1 definition in two important respects. First, in 2000 came the signing of the Palermo Protocol, the United Nations instrument related to trafficking which set out a definition for trafficking in persons which explicitly mentioned slavery. This was followed, in 2002, by the coming into force of the Statute of the International Criminal Court which established the crime against humanity of ‘enslavement’. The crime was defined, in substance, in the same terms as slavery, as: ‘the exercise of any or all of the powers attaching to the right of ownership over a person and includes the exercise of such power in the course of trafficking in persons, in particular women and children’.
We might suppose then, that recent years would have seen some interpretative consensus on the definition of slavery around the terms of the 1926 Convention, but the opportunities occasioned by 2000 and 2002 proved to be very much a false dawn. In fact the result in legal contexts has been an even more blurred understanding of what counts as slavery. This is well illustrated by the first two international judgments to appear at the start of the new millenium, which interpret the provisions of the 1926 definition and come to diametrically opposite understanding of what slavery means. In 2003, the Appeals Chamber of the Yugoslav Tribunal, in the Kunarac case—related to detention centres in the region of Foca, Bosnia and Herzegovina, where women where enslaved in what were in essence rape camps—sustained ‘the chief thesis of the Trial Chamber that the traditional concept of slavery, as defined in the 1926 Slavery Convention […] has evolved to encompass various contemporary forms of slavery which are also based on the exercise of any or all of the powers attaching to the right of ownership’.Footnote 23 The Appeals Chamber of the International Criminal Tribunal for the former Yugoslavia went on to say ‘that the law does not know of a “right of ownership over a person”. Article 1(1) of the 1926 Slavery Convention speaks more guardedly “of a person over whom any or all of the powers attaching to the right of ownership are exercised”. That language is to be preferred’.Footnote 24 This determination is to be contrasted with that of the European Court of Human Rights which, in 2005, considered that the 1926 definition was only applicable where there was a de jure right of ownership over a person. In the Siliadin case, which dealt with the fate of a Togolese girl who was held in servitude as a live-in domestic worker in Paris, the European Court stated that the definition of slavery ‘corresponds to the “classic” meaning of slavery as it was practiced for centuries. Although the applicant was, in the instant case, clearly deprived of her personal autonomy, the evidence does not suggest that she was held in slavery in the proper sense, in other words that [the perpetrators] exercised a genuine right of legal ownership over her, thus reducing her to the status of an “object”’.Footnote 25
In an effort to resolve this kind of interpretative deadlock, many have turned to the foremost expert on contemporary slavery, Kevin Bales. His sociologically-based understanding of slavery pervades recent legal discourse, and is taken to provide a normative basis for modern slavery. Thus, for instance, in a 2009 Human Rights Quarterly piece entitled ‘Defining Contemporary Forms of Slavery: Updating a Venerable NGO’, Claude Welch states that ‘international actors have struggled to find an adequate definition of “slavery”’. Under the heading ‘Forging a New Definition of “Slavery”’, Welch juxtaposes what he terms the ‘“New” forms of slavery manifest in the servitudes set out in the 1956 Supplementary Convention with “Kevin Bales’ New Slavery”’.Footnote 26 In so doing, Welch seeks to argue that the 1926 and 1956 Conventions fail to adequately address contemporary forms of slavery, and he ends his consideration by stating that ‘the struggle to eradicate all forms of slavery […] continues’.Footnote 27 As for Bales's understanding of slavery, it has the benefit of clearly setting out markers which allow for individuals to identify the sociological attributes of slavery. In his 2009 The Slave Next Door, Bales along with co-author Ron Soodalter, states that:
Slavery has been defined in various ways, but there are three essential criteria from knowing if someone is a slave. The first is the complete control of one person by another, through the use of violence – both physical and psychological. The second – hard labour for little or no pay – clearly applies as well. […] The third criterion is economic exploitation – making a profit for the slaveholder.Footnote 28
This conception is very helpful, but it does not follow, as Welch suggests, that the 1926 definition fails to address contemporary forms of slavery. Indeed, our argument in the next section will be exactly the contrary.
As we begin to make this argument, it is worth noting first that, since Kunarac, in four notable instances the cases of the past few years have shown a trend towards accepting that the 1926 definition has contemporary relevance. In the 2007 Brima case before the Special Court for Sierra Leone; in 2008, the ECOWAS Community Court of Justice found Niger, in Mani Karoau, in violation of prohibition against slavery as set out in the African Charter on Human and Peoples’ Rights; also in 2008, the High Court of Australia in Tang, confirmed a violation of the provision of the Australian Criminal Code which incorporates the 1926 definition; and finally in 2010, the European Court of Human Rights, in Rantsev, moved away from its earlier pronouncement in Siliadin.Footnote 29
As for the European Court of Human Rights, while it repeated its determination from the Siliadin case, it turned also to Kunarac, stating that the Yugoslav Tribunal had ‘concluded that the traditional concept of “slavery” has evolved to encompass various contemporary forms of slavery based on the exercise of any or all of the powers attaching to the right of ownership’. It followed this by leaving behind the notion that the definition of slavery was applicable only in de jure situations, finding ‘that trafficking in human beings, by its very nature and aim of exploitation, is based on the exercise of powers attaching to the right of ownership’.Footnote 30 While this acknowledgment by the European Court of Human Rights is welcomed, the most thorough consideration of the 1926 definition in these four determinations is found in Tang, a decision of the High Court of Australia (the Australian supreme court).
In Tang, the High Court found, on appeal, a Melbourne brothel owner guilty of five counts of possessing a slave and five counts of ‘intentionally exercising over a slave a power attaching to the right of ownership, namely the power to use’.Footnote 31 In turning to its consideration of slavery, the Court dismantled the 1926 definition by first considering the distinction between ‘status or condition’. The Court stated that: ‘Status is a legal concept’. With regard to ‘condition’, the Court determined that ‘the evident purpose of the reference to “condition” was to cover slavery de facto as well as de jure’. It went on to explain the application of the 1926 definition to de facto situations, stating that:
In its application to the de facto condition, as distinct from the de jure status, of slavery, the definition was addressing the exercise over a person of powers of the kind that attached to the right of ownership when the legal status was possible; not necessarily all of those powers, but any or all of them.Footnote 32
The Court then turned to consider what those ‘powers attaching to the right of ownership’ might constitute, and pointed to a 1953 Memorandum by the United Nations Secretary-General. Speaking specifically to the powers set out by the Secretary-General,Footnote 33 the High Court determined that ‘on the evidence it was open to the jury to conclude that each of the complainants was made an object of purchase […]; that, for the duration of the contracts, the owners had a capacity to use the complainants and the complainants’ labour in a substantially unrestricted manner; and that the owners were entitled to the fruits of the complainants’ labour without commensurate compensation’.Footnote 34
This approach of the Court in Tang, according to which the determination as to whether slavery exists is made through analysis of facts amounting to the exercise of ‘powers attaching to the right of ownership’, is in keeping with a plain reading of the 1926 definition, and also brings to that definition the substance necessary to give it workable contemporary relevance.
What we lack, so far, is a theoretical basis for this definition, one which can explain its normative underpinnings and its practical operation, and so convince courts, commentators, and anyone interested in ideas of slavery, that in Article 1 we have a useful authoritative reference point which is not so wide as to render the very notion of slavery meaningless. In explaining its approach in Tang, the High Court did not feel compelled to go beyond the pronouncement of United Nations Secretary-General; but that determination is more than 50 years old and sets out examples of ‘powers attaching to the right of ownership’ without seeking to give an overall understanding of that construct. The construct, though, is very familiar to property lawyers. Indeed, much of our jurisprudence on ownership relies on exactly the same differentiation between the legal right ‘ownership’ and the various powers and privileges which might be consequent on that right. Accordingly, to seek to understand the parameters of what constitutes slavery in a contemporary context, we find ourselves in unusual and slightly counterintuitive position of arguing that reference to property perspectives can bring clarity and purchase to the definition of slavery in a world which has long since abolished private property rights in respect of persons. This contention we make in the next section before arguing further that, as rendered by property perspectives, the 1926 definition provides a powerful way forward for identifying and prosecuting modern cases of slavery.
II. APPLYING PROPERTY PERSPECTIVES TO THE DEFINITION OF SLAVERY
In seeking to understand and explain ownership, jurists old and new have laid their emphasis on what the owner of something is entitled to have, do or expect in respect of that thing. Formulations vary, but it is fairly uncontroversial to suggest that the owner of a tangible thing might possess and use her thing; she might make decisions about how it can be used by others; she might enjoy income produced by the thing, sell it to realize its capital value, or simply give it away, including to those who survive her as her heirs. We might also say that she is entitled to a certain measure of security. We would not expect her thing to be taken from her without cause, whether by a fellow citizen or the government; and correlatively we would expect her ownership to endure until such time as she decided to stop being the owner, whether by transferring the thing away, or by consuming or destroying it altogether.Footnote 35
Thinking about ownership in this way, we build up a picture of an ‘owner’ as the person in charge of a given resource. We see very clearly her freedom in controlling the thing in question, or, as it has been put recently, in ‘setting the agenda’ for the use of the thing.Footnote 36 This picture is helpful, and we can use it instrumentally to explain the meaning of Article 1. Specifically, without necessarily agreeing with the details of any single formulation of the content of ownership, or committing ourselves irrevocably to an overarching conception of property,Footnote 37 we can engage such property narrative to substantiate the kinds of exploitation against which Article 1 sets it face. So we dwell for a moment on the various rights, powers, liberties or immunities which might be exercised by the owner of something tangible.Footnote 38 This will lay the foundation for a crucial claim made in the next section, which very closely resonates with the reasoning of the High Court in Tang: it remains possible today for the same kinds of control to be exercised in respect of human beings, even though we have abolished property in persons.
The most basic feature of ownership is possession.Footnote 39 If some person A is owner of some thing, she is entitled to have it, and to demand that others do not interfere with her control of it, at least not without her permission. To safeguard A's position, and specifically to restrain such interferences with her possession, legal systems recognize that everyone else in the world is under a legal duty not to interfere with A's control of the thing.Footnote 40A has access to courts and to formal legal mechanisms to enforce this duty, and ultimately she might bring proceedings against a specific defendant, say B, accused of some act of interference.Footnote 41 So A might sue B for return of the thing or its value, and have a judgment against B enforced by the appropriate authorities. In lawyerly language, A has a claim-right against B, as will she also have a claim-right against anyone else who breaches the duty not to interfere with her use of the thing.
The potential for this kind of claim-right is one of the defining characteristics of private law ownership.Footnote 42 In this respect, in law ownership operates negatively, preserving a sphere of freedom for A in dealing with a resource by keeping others from interfering with her control.Footnote 43 Beyond this negative safeguard, the content of A's sphere of freedom is not precisely defined, nor is it achieved by an exactly enumerated list of privileges. A might continue to possess her thing or make whatever use of it she likes just because she is free to do so.Footnote 44 So if A is the owner of a bike, we say A has a liberty to ride her bike, to paint it, lock it up, display it an as ornament, and put it to all manner of other possible uses. Correlatively to this liberty, no one else in the world has the right (B has no right) to prevent A from engaging in the any of these uses, nor any claim to make the decisions in her stead.
If possession and use are secured by claim-rights and liberties, other features of ownership are manifested in the implementation of legal constructions properly called powers. In law a power generally refers to the ability of the holder to bring about changes in the legal relations prevailing between herself and others. In the context of ownership there are some obvious examples. An owner has the power to licence what would otherwise be a trespass. By giving B permission to ride her bike, A can amend the legal relations prevailing between herself and B such that B breaches no duty of non-interference with the bike by taking it for a spin. A can also change the relations more permanently. She has the power to transfer her thing to B, by giving or selling it, in consequence of which the rest of the world now owes a duty of non-interference to B rather than A.
Some features of ownership consist in immunities, that is, in the absence of powers in others to disrupt A's ownership of her thing or it to an end. One manifestation of this is that generally ownership is considered to be of potentially limitless duration, such that it will not (normally) come to an end (or ‘determine’) on the occurrence of some future event (as is the case with some other kinds of property interests, for example, leases of land). Another good example is provided by rules against unjustified expropriation. This immunity preserves a private sphere of autonomy in dealing with assets, and ensures that A might safely make plans for her life on the basis that her given allocation of things will not be disturbed without some due process (such as the procedures for bankruptcy, execution for debt, justified compulsory acquisition). In the same way, it ensures that, apart from the operation of such process, transfer or alienation of ownership occurs only with the consent of the present owner, and again this preserves the autonomy of the owner as director of a particular resource.
Private law ownership of human beings is today a ‘legal impossibility’,Footnote 45 but the discussion above will allow us to be precise about the technical effect of this abolition, and consequentially will offer a warning: notwithstanding the technical absence of property rights, it remains possible for the same kinds of control to pertain in respect of persons in ways which continue to be objectionable. First, the abolition of property in respect of human beings means obviously that some person (the slave-holder: call him SH) is no longer permitted to hold a legally enforceable claim-right in respect of another person under his control (call that other S), and correspondingly no court will entertain litigation in respect of interferences with a person-controlled. Recall, though, that the broad function of claim rights was to preserve a sphere of freedom in using the thing in question. In abolishing claim-rights in respect of persons, we do not necessarily do enough to disrupt these spheres of freedom. For sure, we can say that those freedoms are no longer guaranteed by law, in that law no longer demarcates and preserves a realm of liberty-in-dealing with persons in the manner achieved negatively through the operation of claim-rights. In other words, we can say that the law will no longer protect SH's use of S. We can point further to the fact that, strictly, SH can never now have a liberty to use S, because S will have countervailing rights to resist use and interference with her person (for example, through the civil law dealing with assault, battery and false imprisonment). But nonetheless, in the general circumstances of exploitation, and notwithstanding the legal protection available in theory to S and the absence of any such protection for SH, it remains possible in fact for one person to exercise over another sufficiently close analogues of the liberty-of-dealing associated with ownership. So SH might continue to control and use S by exploiting her services or labour, as where she is used for sexual gratification or to provide domestic services. S might be moved from one place to another; her ties to a particular culture, society or religion may be severed against her will; her means of access to state support or authority may be curtailed. And eventually, when her purpose is exhausted, she might be disposed or discarded, as a thing no longer worth keeping. These circumstances which might be occasioned by SH, whilst they cannot formally be regarded as flowing from the exercise of any liberty associated with ownership, are nevertheless sufficiently similar, and in many respects exactly the same, as consequences which may follow from de jure ownership.
Something similar obtains in the case of powers. Where ownership of persons has been abolished, SH does not have, say, the power to licence what would otherwise be a trespass, but there will be sufficiently close factual analogues. This is very clearly so in the case of sex slavery, and also in slavery for labour, where SH might, for example, direct a manager to discipline workers with force or physical threats, or ‘lend’ to another the services of S for a period of time.Footnote 46 In the same way, while SH certainly cannot sell S to another, and cannot in that sense bring about any formal change in the legal relations prevailing between that person and the rest of the world in respect of S, we will easily find real examples of people being traded for money, or otherwise made the subject of a transaction.Footnote 47 Similarly, while SH lacks the power to bequeath S to anyone, nonetheless there will remain local customs, procedures and practices which will serve to bring about the transfer of control of a person to an heir or successor, as where a woman, on the death of her husband, is deemed to be inherited by another person; or where on the death of a person, a subordinate status passes from her to a member of a successive generation.
De facto manifestations of liberties and powers to use, manage, profit from, and discard a person seem fairly central examples of the kinds of control of persons that we might find in the modern world, but the immunities associated with ownership have resonance here too. As to the potentially limitless duration of ownership, one of the cardinal features of slavery is that, from the perspective of the slave, the control she experiences is indefinite, and it is outwith her will to bring it to an end.Footnote 48 As to security of holding, whilst at first glance it seems difficult to transfer this incident of ownership to the context of persons,Footnote 49 again we are likely to find sufficiently close factual analogues. In particular, since expropriation of property is often considered to be politically legitimate within well-defined terms (as in the cases of bankruptcy or compulsory acquisition by the state), practically there will be instances of States operating to disrupt exploitative relationships of control (forced labour markets, prostitution rings) which can operate as analogies to cases of justified expropriation, and in so doing highlight the persistence of modern de facto control relations. Moreover, where a state government does not take steps to disrupt such activities, arguably this serves to create de facto security of holdings in respect of persons, reinforcing the security of such illegitimate relations.
Accepting generally that the phenomena of ownership might persist in this way allows us effectively to endorse the decision of the High Court of Australia in Tang, by observing that the de jure abolition of formal slavery relations offers no guarantee against their persistence in fact. Moreover, it brings normative purchase to the Article 1 definition, since very clearly the exercise in fact of these kinds of control in respect of a person cuts to the heart of the objection to slavery. Slavery is wrong when the fundamental assumption of our political philosophy articulates the basic freedom and equality of persons. In objecting to slavery, we object principally to the subjection of one person to another, to the inroads on the freedom of that person that come when she is controlled or used in ways inconsistent with her choosing and inimical to her plans for life. It follows that any enquiry to the existence of slavery should be concerned first with the real, factual existence of such constraints. As Hayne J put it in Tang, the concern should be to identify ‘a state of affairs in which there is the complete subjection of’ one person by another.Footnote 50 A focus on the phenomena of ownership helps with this because de facto exercise of the powers attaching to ownership in respect of a person serves to indicate real restrictions on that person's freedom.
Perhaps another way of expressing all of this is to say that there is no very great gulf between legal and factual senses of ownership. Ownership is concerned with control. Where the objects of control are ordinary objects of property rights (bikes, cars, horses, coats) the great majority of the owner's exercises of control will nevertheless occur without the specific intervention of the law. Ultimately the owner can call on the law to affirm her position vis-à-vis the thing by taking legal proceedings against an interferer, but most of her actions will be factual exercises of a liberty to control, or invocations of powers to alter existing legal relations. All things considered, this position is not so far removed from contemporary cases involving the subjection of persons. In the modern world it remains possible to exercise exactly the same kinds of control in respect of a person, to behave as if exercising a power and create the factual consequences of so doing. Since such exercises of control remain objectionable, it makes sense to continue to refer to the language of ownership in identifying slavery, as much the greater part of ownership's import continues to have resonance in the slavery context.
III. IDENTIFYING MODERN CASES OF SLAVERY
Accordingly, the 1926 definition of slavery is useful and important because it opens up a substantive understanding of modern slavery. Invoking the conception of ownership discussed above, it points us away from the necessity of formal slavery relations towards the substance of the problem: exploitative relationships where one person is controlled by another as if she were a thing. In so doing, it correlates the legal definition of slavery with our principal objections to slavery as a social institution, removing the need for any wider understanding of slavery which goes beyond a plain text reading of its provisions, whilst at the same time engaging in a systematic way the concern to address modern forms of subjection which count as slavery. Thus, Article 1 has demonstrable contemporary relevance, and fully deserves its status as the starting point for any enquiry as to whether a given practice should be considered slavery. To help courts and prosecutors (or anyone else interested) to proceed from here and identify modern cases of slavery, we can usefully make two further interrelated observations on the text of the 1926 definition, and then unpack a little more what it means to exercise ‘the powers attaching to the right of ownership’ in respect of a human being.
First, it will not have escaped the keen reader that Article 1 speaks in terms of ‘powers attaching to’ ownership, whereas we have seen that ownership employs a broader range of legal constructs only some of which can accurately be considered as powers. We think it is not necessary to read Article 1 literally and restrictively as being concerned only with the exercise of powers or their factual equivalents, as opposed to factual equivalents of the wider phenomena of ownership (claim-rights, liberties and immunities). It might be enough to convince on this point that the legal discourse which conceives of legal rights in these constitutive elements of claim-rights, liberties, powers, and immunities was still very novel in the 1920s,Footnote 51 at the time when Article 1 was drafted, and nowhere near as well established as the general conceptual enterprise of understanding ownership in terms of what the owner was entitled to do or expect.Footnote 52 More significantly though, as we have tried to show above, and as the Court made clear in Tang, the point about slavery is that it results in the total subjection of one person to another. Powers are not the only incidents of ownership which if exercised de facto will contribute to subjection. Liberties of possession and use also figure prominently. Indeed, if we take possession to mean the maintenance of effective control, we must take possession as an essential, foundational aspect of the kind of subjection envisaged in Tang. So we cannot, and need not confine Article 1 to a strict reading of power. For sure that definition embraces analogues of some behaviours correctly thought of as the exercise of a power, but it also embraces the exercise of liberties, and analogues of immunities too. As such, we should read the phrase ‘powers attaching to the right of ownership’ only as engaging a general conceptual tool which seeks to understand ownership in terms of its incidents, and not in any narrower sense.
Secondly, the 1926 definition recites that the exercise of ‘any or all’ of the ‘powers attaching’ indicates the presence of a slavery relation. It would seem to follow on a literal reading that whenever we observe in practice the exercise of any one incident of ownership in respect of a person, this serves to indicate that slavery exists. However, this will often be insufficient or problematic. To take a trivial and obvious example, employers routinely expect to make decisions about the tasks and activities of their employees, and might reasonably expect to be entitled to the profits of their labour, subject to payment of a (fair) wage; but only with invective would we refer to the employment relation as slavery, and certainly we would not mean it to fall within the scope of Article 1 of the 1926 Convention. Similarly a government might detain or restrict the movements of identified individuals, and at least some of the time we can recognize these practices as being in pursuit of legitimate goals. At first blush, then, it does not look to be enough in identifying slavery that we point to the existence of any of the powers attaching to the right of ownership, though that seems to be what a literal reading of Article 1(1) intends. We need to find something more general to say about the effect of the incidents of ownership in the slavery context that can help us distinguish sound manifestations from unsound ones, and so explain why Article 1 adopts the language of ‘any or all’.
When jurists approach the matter of elucidating and explaining ownership, they do so in the context of a background understanding that ownership implies a particular relation of control (this explains why possession is considered the foundational incident of ownership). For Roman lawyers, the owner was dominus, master of the thing owned,Footnote 53 and similarly, when Blackstone spoke of ‘property’, he supposed a despotic relation of control between person and thing.Footnote 54 More recently, as we saw earlier, Larissa Katz has referred to the owner as ‘the exclusive agenda-setter for the owned thing’.Footnote 55 This is a particularly helpful conception because it carries the idea of the owner's empowerment to deal with and make decisions about the thing, and also of the thing's subjection to the will of the owner. That subjection need not be total: others might have a part to play in determining the use of a thing on particular occasions;Footnote 56 but ultimately the owner is the person calling the shots, the person who gets to manipulate and control the resource to meet his own ends.
It seems to follow that, when we seek to identify slavery, we should look for manifestations of the powers attaching to ownership which occur against the background of this kind of control relationship between persons. This entails, at a minimum, that slavery occurs where one person controls another as he would control a thing possessed; or to put it more shortly, that slavery entails control of a person tantamount to possession. This understanding resonates with the conception of ownership engaged by Article 1, but also provides descriptive and normative explanations for the ‘any or all’ criterion. Descriptively, control will almost certainly be a factual prerequisite for any de facto exercise of the powers attaching to ownership. A person can only be sold, lent, used or discarded if first she is within the realm of control of the seller, user, lender or discarder. Absent the presence of such control, it would largely be meaningless to speak of SH entering into any kind of transaction in respect of S, or putting her to any use: such things just would not be possible in fact. Further, if SH must first control S tantamount to possessing her, we imbue with normative significance any exercise of a power attaching to the right of ownership in respect of S. Selling, using, lending or discarding S testifies to SH's control, and to S's subjection. It is on the wrongness of these that we premise our objection to slavery.
If slavery entails control tantamount to possession, an important corollary is that it will not necessarily constitute slavery where one person exercises any one incidental power of ownership in circumstances where such control is absent. This accounts, for example, for the situation where the employer makes routine management decisions about her worker; but it might also explain more difficult situations, where though there has been some control exercised in a manner which is exploitive, yet this control amounts to something less than possession and thus is not slavery but, say, forced labour. An example of such an instance would be the case of a textile employer exacting forced labour but not controlling other elements of a worker's life, thus failing short of the threshold of control tantamount to possession.Footnote 57
Having established the significance of control as being foundational to slavery, in that it makes possible the de facto exercise of further powers attaching to ownership and articulates our objection to the same, the next step is to be more specific about how such exercises might manifest themselves in situations of modern slavery. Both the United Nations Secretary-General and the legislation of the International Criminal Court provide certain examples but, as yet, no systematic attention has been directed to the question of what would constitute exercise of such powers against a human being.Footnote 58 Perhaps the most intuitive exercise of such a power would be in the buying or selling of a person.Footnote 59 In such a transaction, as in other similar transfers of a person, be it through exchange, barter or gift, it becomes obvious that the foundational element of possession is required for slavery to be present. It is only through such control, often established through violence or coercion that a person is made amenable to being the subject of a transaction. Once a person has become the subject of a transaction, that transaction will generate manifold facts which can testify to the existence of slavery, from the revenues and funds associated with any exchange, which approximate to the realization of S's capital value as a thing, to the physical changes in S's custody and any force applied to bring them about.
Less intuitive, though very present in contemporary practice is the transferring of a person to an heir or successor, in situations which constitute slavery. Most evident are cases of inheritance, where a woman, on the death of her husband, is deemed to be inherited by the family of the former husband.Footnote 60 This practice is more common than descent-based slavery, which we will still find in Mauritania or Niger, where there continue to be instances of the conveying of a slave to a successive generation, such as from a father to his son.Footnote 61 In the case of inheritance, as in the case of the purchasing of a person, the expectation is that benefit will be derived from the service or labour of a slave.
This same benefit underlies our next point as it is easy to suppose, given control tantamount to possession, that a person can still be used as if she was a thing, and in such use of a person we see an analogue of the function of ownership. This will be manifest where a person is used for sexual gratification, or to provide a service or labour for little or no pay. Such was the case, for instance, in Sierra Leone where, during the 1991–2002 Civil War, women were abducted and designated as ‘bush wives’ by the soldiers who took possession of them and forced them under threat of death to cook, clean, and porter.Footnote 62
Managing of the use of a person will constitute exercise of a power attaching to the right of ownership amounting to slavery, where the background relationship of control tantamount to possession is present, as where, for example, the day-to-day management of so-called ‘captive communities’ within Bolivian cattle ranches is delegated by the owner of such estates to over-seers/manager.Footnote 63 At a more abstract level, such management of the use of a person will also be manifest in certain steps taken to consolidate control tantamount to possession. Examples include the imposition of a new place of residence, language, or religion, or forcing marriage so as to sever previous social bonds.
In like manner, profiting from the use of a person constitutes exercise of a power attaching to the right of ownership, given a background relationship of control tantamount to possession. Such profit might be gained through mortgaging a person, lending her out for profit, or using her as collateral: indeed, in principle, it may include making money or deriving any other kind of income or benefit from her use in any way whatsoever. We can draw a practical example from an instance of an agricultural worker in a situation of slavery, where the profit from the picking of a crop is taken by another, whether in the form of unpaid wages or money derived from the selling of the harvest.Footnote 64
Closely associated with the concept of use would be the power to dispose of one's property. This leads us to suggest that a further exercise of a power attaching to the right of ownership would be manifest in the mistreatment or neglect of that person, inasmuch as it shows that she is being treated as a disposable person.Footnote 65 Any such disposal, mistreatment or neglect of a person will fall within the parameters of slavery where control tantamount to possession is present. In such instances, mistreatment or neglect leads to the physical or psychological exhaustion of a person, and ultimately may lead to her destruction if it persists. Such mistreatment might be either calculated or indiscriminate, and would impose such demands as to severely curtail the capacity of her body to sustain itself or function effectively. Examples of situations of slavery would be cases where children are, after a certain period in a mine or begging on the street, no longer deemed effective and are thus abandoned.Footnote 66
Finally, as we said earlier, immunity from expropriation, or ‘security of holding’ is normally associated with ownership. Generally this means that a State has no power to interfere with private property, but sometimes it will be permitted so to act where prescribed by law, in situations of public interest. In the context of modern slavery, we will find analogies to such permissive disruption of property rights, which can serve to remind us of the persistence of slavery relations. A good example can be found in the jurisprudence of the European Court of Human Rights, which accepts that States of the Council of Europe have a number of far-reaching positive obligations to disrupt property relationships in persons, of which the minimum, which is accepted by all States, is simply to bring about the end of either the status or the condition of a person over whom any or all of the powers attaching to the right of ownership are exercised.Footnote 67
It is thus possible to construct the exercise of powers attaching to the right of ownership in respect of a person. They include (though we do not intend this list to be exhaustive) buying or selling a person; transferring a person to an heir or successor; using a person; managing the use of a person; profiting from the use of a person; disposal, mistreatment or neglect of a person; and as analogous to the immunity from expropriation, the States’ positive obligation to bring about, by various means, the cessation of slavery. Control tantamount to possession is a minimum condition for their exercise, but given such control, the exercise of any or all of these powers speaks to the condition of slavery, and thus, possession, as a basic feature of ownership, serves to identify slavery in practice.Footnote 68
It is no answer to say that formal relations of slavery have been abolished and with them the notion that private property rights might obtain in respect of human beings; because it is remains possible for subjection to persist in fact. Accordingly, attention must be directed not to the technical exercise of any power but rather to the occurrence of facts analogous to those produced by the exercise of power, like the fact that SH received a sum of money from B in exchange for delivering S into the control of the B. In other words, we must move beyond the form of slavery and look to the substance. In our view, this is the vision and understanding of slavery which is entailed by the conception of ownership employed in Article 1. It is encouraging to note in this conclusion that, in 1953, the United Nations Secretary-General was attuned to making such a distinction between form and substance, stating that the 1926 Slavery Convention extends to ‘all institutions or practices, whether or not designated as “slavery”, provided that, as stated in Article 1 of the Convention, “any or all of the powers attaching to the right of ownership are exercised” over a person in these institutions or practice’.Footnote 69
Notice that, in this sense, the definition in Article 1 is expansive rather than restrictive, and appropriately configured to address the changing circumstances of modern slavery. It is targeted against all forms of exploitation which satisfy the crucial requirement for slavery: that a person is being controlled as if she were a thing possessed. This is the core meaning of slavery, and in postulating such a core, Article 1 avoids the pitfalls of the spiralling interpretations of slavery that have characterized much of the discourse from the League of Nations era on through much of the twentieth century.
It follows that whether or not slavery exists in practice in a given case will hinge on a factual determination that there has been de facto exercise of one or more of the powers attaching to the right of ownership in circumstances where control tantamount to possession is present. The High Court in Tang adopted such a substantive approach, and prosecutors and courts considering future cases of slavery should take its lead, using the rhetoric of ‘powers attaching to the right of ownership’ to scrutinize the factual circumstances of subjugation. Where any or all of these powers is exercised in circumstances of control tantamount to possession, slavery exists. Where such control is not present, even though the behaviours in view are exploitative and certainly wrong, yet they are not slavery, inasmuch as they fail to exhibit the core meaning of slavery. Importantly, this accounts for a distinction between slavery and those lesser servitudes also recognized by international law, such as forced labour, and the 1956 Supplementary Convention's ‘institutions and practices similar to slavery’, namely, debt bondage, serfdom, servile marriage, and child trafficking. While in principle, a particular circumstance of exploitation might justify a finding of slavery or a finding of one of these lesser servitudes, the appropriate manner to proceed is to address the substance of the situation rather than the form, by asking whether there has been de facto exercise of any or all of the powers attaching to the right of ownership.Footnote 70 Where this inquiry is answered affirmatively in circumstances disclosing control tantamount to possession, the more serious condition of slavery is present. Where the inquiry brings a negative response, prosecutors and courts can appropriately move to consider the legal definition of the lesser servitude which corresponds to the particular circumstance in question.
In this respect, consider the difference between slavery and the lesser servitude of forced labour. The 1926 Convention, for its part, recognizes that forced or compulsory labour can develop ‘into conditions analogous to slavery’. As such, forced labour, while meeting its legal definition as set out in the 1930 Forced Labour Convention (‘all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily’Footnote 71) may also amount to slavery in law, if such labour transpires in a situation where the powers attaching to the right of ownership are exercised. Just as the intentional infliction of serious harm upon a person may constitute both assault and, if death ensues, murder, so the circumstances of forced labour will amount to slavery when the critical element of the latter is present on the facts. In a situation where a clerk is forced to accept less than minimum wage under threat of dismissal, this constitutes forced labour; however the threshold of control would have to be much higher for such an instance to constitute slavery. As we have sought to demonstrate through our reading of the 1926 definition, that threshold would be where a person exercises control tantamount to possession.
In practical terms, the distinction between forced labour and slavery might be adduced from the facts of the 2007 Brima case before the Special Court of Sierra Leone.Footnote 72 While the Trial Chamber did not make this distinction in finding that individuals were enslaved and put to work in the Tongo diamond fields, the facts might be understood as comprising certain cases where forced labour was compelled which did not meet the threshold of enslavement. This is so, as in some cases, individuals were captured by rebel forces in Sierra Leone, undressed, tied together and brought to the mines ‘where they were forced to work at gunpoint’.Footnote 73 In such situations, the control exercised was tantamount to possession, as the individuals were at the complete mercy of their captors. Yet, in other cases a witness testified that other miners were compelled, under threat of violence, to spend two ‘government days’ a week mining diamonds. For these individuals, while they were not enslaved, as they were free after their working days to return to their homes and to work on other days at the mine for their own profit; however they were subject to forced labour.Footnote 74
IV. CONCLUSION
Article 1 of the 1926 Slavery Convention has long been accepted as the authoritative source on what counts as slavery in law, but subsequent interpretations of its provisions, premised on agenda-driven desires to bring modern forms of exploitation within the purview of slavery, have lacked consistency and resulted in large measures of confusion. The chief shortcoming has been a failure to engage at all with the ownership construct employed by the text of Article 1. Our purpose in bringing forward a theoretical analysis of the definition of slavery, viewed through the lens of this property paradigm, has been to explain the normative underpinnings of Article 1 and its practical operation, so as to deliver a definition of slavery with a core settled meaning and demonstrable contemporary relevance.
The task for States is to bring this machinery to bear on the behaviours and practices of modern subjugation. Courts should be alive to the import of Article 1 when determining whether slavery exists in a given case, and in its provisions they have a clear structure for how to proceed when making determinations of this kind. For the rest of us, Article 1 should serve as a reminder that, whilst formal slavery relations have long since been abolished, and in that sense ‘legal slavery’ is no longer possible, international law is not neutral on the matter, and neither are its provisions benign. Slavery still exists in law where any or all of the powers attaching to the right of ownership are exercised in respect of a person subject to control tantamount to possession; slavery still exists in law where a person is controlled as if she were a thing. Its manifestation as enslavement before the International Criminal Court, its prohibition in all major international human rights treaties, and its inclusion in both the United Nations and European anti-trafficking conventions, provides the foundation which, when coupled with the legal certainty of its meaning in law, allows human subjection to be recognized as a contemporary wrong to which international law speaks.
