Commentary on the Safety of Rwanda (Asylum and Immigration) Act (“RA”), which is shortly to receive Royal Assent, has concentrated principally on its deeming of Rwanda as a safe country whilst ousting the supervision of courts. This post considers a separate issue – section 4 of the Act as it applies to victims of slavery (“VOS”). Section 4 provides a carve out from the Act’s deeming provisions where the Home Secretary considers Rwanda is unsafe for an individual “based on compelling evidence relating specifically to their particular individual circumstances”. It also provides courts with a power of review of that question.  

This post argues that, read in the light of the common law constitutional prohibition of slavery (“POS”), s.4 should prevent all suspected and confirmed victims of slavery from being removed against their will to Rwanda without, at the least, a detailed assessment of their specific risks of re-trafficking there.

Modern Slavery victims and the Rwanda Act

Last week the House of Commons rejected a Lords amendment tabled by Baroness Butler-Sloss which would have prevented suspected victims of modern slavery (those with reasonable grounds National Referral Mechanism decisions) from being removed without their consent, and prevented confirmed victims (those with conclusive grounds NRM decisions) from being removed without a “decision-maker considering whether such removal would negatively affect the physical health, mental health or safety of that person, including in particular the risk of re-trafficking.” The Commons initially rejected the amendment on the basis that “provisions for modern slavery and human trafficking victims are set out in existing legislation.” When the Lords insisted, the Commons inserted the annual reporting obligation which appears in the Act. The reporting function in the RA will not prevent suspected and confirmed VOS being removed against their consent.

The effect of the lack of protection for VOS in the RA is dependent on provisions of the Illegal Migration Act 2023 (“IMA”) which, at the time of writing, have not yet been brought into force. Once they are, the combination of the IMA and RA will decimate the domestic law framework to protect victims of slavery. The SSHD will be under a duty, under s 2 of the 2023 Act, to remove as soon as reasonably practicable all suspected and confirmed victims of slavery who entered the UK without leave (other than those who came directly from a country in which the person’s life and liberty were threatened by reason of their race, religion, nationality, membership of a particular social group or political opinion) after 20 July 2023 (unless that date is amended by regulations under s.3 IMA). The duty will apply regardless of reasonable grounds/conclusive grounds status. The current systems under which a recovery period must be granted to suspected victims, and temporary leave to remain may be granted to confirmed victims will not apply (unless the victim falls into a very limited exception relating to criminal investigation and proceedings)_- IMA 2023, s 22.

These changes, and the removal of suspected and confirmed victims of slavery to Rwanda, will be in breach of the UK’s international law obligations under the European Convention against Trafficking (ECAT), currently incorporated under the Nationality and Borders Act 2022, and almost inevitably breach Article 4 ECHR. However, when they come into force, the changes will also unincorporate ECAT, disapply the obligation to interpret the IMA and RA compatibly with ECHR rights under s.3 HRA, and (as far as s.4 RA is concerned) disapply the obligation under s.6 HRA on the SSHD to act compatibly with ECHR rights (although declarations of incompatibility under s.4 HRA will remain available). 

With ECHR and ECAT protections denied to victims of slavery in domestic law, the question arises whether they might find protection in the common law. 

Common law constitutional rights

The term common law constitutional rights (“CLCR”) is generally used as an umbrella term for bundles of rights and interpretative obligations which protect core fundamental rights in English law (for instance, engaging the principle of legality). CLCR seemed big in the 2010s: the subject of landmark judgments, judicial journal articles, and academic books and conferences. However, recent Supreme Court judgments on CLCR have told us which rights we don’t have (citizenship; to prevent Government action which facilitates execution abroad), as opposed to which ones we do. Mark Elliott and Kirsty Hughes’ 2020 observation that “we have limited insight into what rights we have, how they work, and what they offer,” therefore, remains acute [p.5].

Interest in CLCR seems to have waned somewhat recently after continual Governmental threats to replace the HRA with a domestic bill of rights and/or to leave the Convention altogether failed to materialise. But rather than repealing the HRA altogether, Parliament has set about quietly dismantling Strasbourg protections on a statute by statute basis (a process referred to by Stuart Wallace as the death of the HRA by a thousand cuts). The IMA and the RA are prime examples. These developments make identifying the existence and scope of CLCR more important than ever. But there is value in undertaking that task in any event, not least as the nature of Parliamentary Sovereignty and common law adjudication make it difficult to predict when the question of judicial recognition of CLCR will arise. As Lord Reed emphasised in his foreword to the book CLCR in 2020 [at vii], with the “implications of the idea of common law constitutional rights still being examined by the courts in a range of situations”, academic contributions have a particularly important role to play in “point[ing] out possible ways forward for courts to steer a course between the Scylla of sterility and passivity, and the Charybdis of instability and uncertainty.”

Analysis of CLCR is a two-step process. Before considering its interpretative effects, it is first necessary to identify a pre-existing fundamental common law right. Thus in the PRCBC appeal Lord Hodge DPSC at [33-34] held that the principle of legality did not apply because the right to British citizenship, unlike the right of access to the court or freedom of expression, has never been recognised by the common law. 

There is no canonical statement in the case law of the criteria for identifying a CLCR. The most detailed proposed methodology is to be found in Lord Sales’ 2016 article ‘Rights and Fundamental Rights in English Law’. It advocated “an approach grounded in history and constitutional tradition”, which is not strictly a “common law method”, but one “of recognition by the courts of constitutional understandings within the polity as a whole” [pp 95-97]. On Lord Sales’ model the range of evidence and argument which may be used in identifying whether a fundamental right is embedded in the constitution includes not just the consistent identification of a particular right or norm over time in case law, but domestic statutes which evidence constitutional understandings, international human rights instruments, and comparative jurisprudence. With the addition of the work of the great common law jurists, this method is consistent with cases in which CLCR arguments have succeeded, including Unison and Belmarsh No.2.

Identifying the Common law Prohibition of Slavery 

The right not to be enslaved certainly sounds like one we should all be able to get behind. In his recent Atkin Lecture, Professor Richard Ekins KC, head of Policy Exchange’s Judicial Power Project and a sceptic of human rights law, supported the notion of it as an absolute human right [p.9]. But despite slavery’s prominent role both in English legal history and in the contemporary human rights law framework, no modern case or text has yet analysed the POS in the common law. 

That is an omission in the learning, because the evidence in support of a common law POS, identified according to Sales’ model, is overwhelming. The jurisprudential foundation of the common law POS goes back to our earliest jurists. Chattel slavery did not exist in England from around the 12^th century, but it gave way to a customary system of villeinage which Bracton, Fortescue and Coke all compared directly to slavery, by adopting the Justinian Code’s notion that slavery was contrary to the law of nature, but authorised by the law of nations. Bracton drew no distinction of gravity between serfdom and slavery. For him, English law knew no degrees of personal freedom: “[w]hoever is bond is just as much a bondsman as any other, no more, no less.”[p.31]. In his Institutes Coke cited Fortescue’s 15^th Century observation that: 

A law is also necessarily adjudged cruel, if it increases servitude and diminishes freedom, for which human nature craves… In considering these matters the laws of England favour liberty in every case [Ch XLII 61, cited in Coke, A Commentary on Littleton, Lib.II, Chap.XI, Section 172].

Villeinage died out and the English law of slavery moved on as the effects of the transatlantic slave trade were felt in domestic courts from the late 1600s. After some prevarication, by the second half of the 18^th century, the common law had turned its face firmly against slavery and the slave trade. In Blackstone’s first 1765 volume of his Commentaries, he adopted the line of authority grounded in the judgments of Chief Justice Holt to the effect that English law did not recognise a right of property in a person (see e.g. Smith v Brown and Cooper 91 ER 566). In his chapter on the Absolute Rights of Individuals he wrote:

… and this spirit of liberty is so deeply implanted in our constitution, and rooted even in our very soil, that a slave or a negro, the moment he lands in England, falls under the protection of the laws, and with regard to all natural rights becomes eo instant a freeman (Commentaries on the Laws of England, First Edition, 1765, Clarendon Press, Oxford, p.411). 

Blackstone’s assertion that slavery is contrary to a constitutional right to liberty provides the jurisprudential background to Lord Mansfield’s famous decision in Somerset v Stewart. Stewart’s pleaded return to Somerset’s writ of habeas corpus relied on a purported right of property to imprison and traffic him to Jamaica. In response Mansfield held, according to the Lofft report:

The state of slavery is of such a nature, that it is incapable of being introduced on any reasons, moral or political; but only positive law, which preserves its force long after the reasons, occasion, and time itself from whence it was created, is erased from memory: it’s so odious, that nothing can be suffered to support it, but positive law.

Somerset has consistently been regarded a case of the highest constitutional principle. Dicey, echoing Blackstone, cited it within the wider principle of ‘personal freedom’, seeing English protections for such “as complete as laws can make them” [An Introduction to the Study of the Law of the Constitution, 10^th Edition, Macmillan, 1965, p.220]. In R (Khawaja) v Secretary of State for the Home Department, Lord Scarman cited it as historic authority for the principle that every person, including migrants, enjoy the equal protection of the common law in the form of habeas corpus [67], a principle further relied upon by Lord Bingham in the House of Lords in the first Belmarsh case [48]. More recently, Green J cited it in the Administrative Court as authority for the proposition that both slavery and treating persons as property were treated as abhorrent to the common law long before the 1833 Slavery Abolition Act [854].

What is the scope of the common law right/s set out in Somerset? Since Fiddes’ 1934 Law Quarterly Review article [(1934) 50 LQR 499] legal historians have questioned whether it decided – at the time – that slavery was prohibited in England. But there is a general consensus that it did decide a slave “could not be forcibly removed from England to… slavery” [Fiddes at 499]. This accords with obiter comments Mansfield himself later that the “determinations [in Somerset] go no further than that the master cannot by force compel him to go out of the kingdom” [R v Inhabitants of Thames Ditton, 99 ER 89]. 

In fact the logic of Somerset – that slavery does not exist in England – and so habeas corpus is available to free a purported slave from detention and to prevent his extradition – does amount to a finding that there is a common law right not to be treated as a slave within the jurisdiction. That is how it was treated in Forbes v Cochrane in 1824 [107 ER 450]. Holroyd J held that the “moment [a slave] puts his foot on the shores of this country, his slavery is at an end…” [463]. Best J agreed but went further. Asserting that slavery is a relation “which has always in British Courts been held to be inconsistent with the constitution of the country”, he held that to detain or use force against a slave in the jurisdiction is a trespass, and went as far as to suggest that if a positive law of slavery was passed in England, judges would be “bound to transgress it” [470]. The picture of the common law’s direct abhorrence of slavery was completed by the 1777 Scottish Court of Session decision in Knight v Wedderburn [1778, Mor 14545] which confirmed that a master could not claim for perpetual service in a servant. 

It is a feature of the POS that the common law turned its face against slavery at least 35 years before Parliament passed the 1807 Slave Trade Act, which was introduced by Wilberforce explicitly in order to protect the rights of vulnerable trafficked people (“the weak against the strong”) on the basis of wider constitutional principles of ”humanity and justice.” The 1807 Act and 1833 Abolition Act (which prohibited slavery in the majority of empire, albeit whilst replacing it with a system of bonded apprenticeship and compensating slave owners), both now repealed, were of historic constitutional significance. Indeed, the offence of Slave Trading under the 1824 Slave Trade Act which derives from the 1807 Act remains on the statute book today, and is – alongside torture – one of the few crimes of universal jurisdiction in England and Wales. Modern historians have rebalanced the traditional British abolitionist narrative to restore historical agency to the enslaved – e.g. by identifying the key role played by Caribbean slave revolts in the abolition of slavery. But for the purposes of recognition of its recognition as a CLCR, what matters is the POS’ wide acceptance over centuries as a consensual English constitutional value. 

During the 19^th century the domestic constitutional prohibition was exported to the international plane. The origins of the contemporary ‘modern slavery’ laws can be traced to the British campaign to end the slave trade through the development of international law (including by slavery being recognised as a crime against the law of nations), which persisted for the best part of the 19^th century, outlasting governments and generations. The anti-slave trade measures in the General Act of the Brussels Conference of 1890 were the explicit foundations for the 1926 League of Nations Slavery Convention, which still provides the definition of slavery in international and domestic English law today. Jenny Martinez has argued that the Courts of Mixed Commission which tried slave trade cases and British success in ending the trade on the international plane during the 19^th Century were responsible for the origin of modern human rights law.

After WWII the UK of course played a core role drafting the Nuremberg Charter, which made enslavement a specified crime against humanity [Jurisdiction and General Principles, Article 6(c)]; and deportation to slave labour a war crime [Article 6(b)]. As well documented elsewhere it also played a core role in drafting the ECHR, which, following the then draft International Covenant on Civil and Political Rights, prohibited slavery, servitude, and forced or compulsory labour under Article 4. 

In the early twentieth century, the international rights under Article 4 were returned to domestic law, and not just via the HRA. The ECtHR’s decisions in Siliadin v France and (after the fact) CN v UK, obliged Parliament to pass the first criminal offence prohibiting slavery (as well as servitude and forced or compulsory labour) in English history under s.71 of the Coroners and Justice Act 2009. After the Strasbourg Court held that human trafficking falls within the scope of Article 4 in Rantsev v Cyprus and Russia in 2010, the Modern Slavery Act 2015 was passed to consolidate existing offences covering the conduct prohibited under Article 4, and to provide additional public law protections to victims. 

The modern English POS therefore includes the conduct covered by slavery, servitude, FCL, and human trafficking. That reflects Parliament’s intention in passing the MSA, it reflects the domestic MSA statutory guidance, the position in European Human Rights law, the position under Inter-American Human Rights law, and in international criminal Law under the crime of enslavement. It was the position recently adopted by the Supreme Court in Basfar v Wong. The right not to be enslaved functions in public, private (via habeas), and criminal law, and is absolute. Neither slavery, nor human trafficking, can be consented to in law [sections 1(5) and 2(2) MSA]. Both are criminal offences punishable by life imprisonment. The prohibition of slavery is a jus cogens norm of customary international law. There is increasing recognition that the prohibition of forced labour is also a norm of customary international law. That was the recent conclusion of the Canadian Supreme Court in Nevsun Resources v Arayya. In 1995 in Kunarac the International Criminal Tribunal for the Former Yugoslavia held that enslavement (when prosecuted as forced labour) is prohibited by customary international law. 

The historical connection between Somerset and Forbes and the English law of modern slavery is completed by recognition that the “abiding” 1926 definition of slavery [Kunarac, 519], and therefore the offence of slavery under s.1 MSA, applies to de facto, as well as de jure slavery. The ECtHR interpreted slavery to require a “genuine legal right of ownership” in Siliadin, but that was plainly an error. Its reasoning was criticised witheringly by Chief Justice Gleeson in the Australian High Court in its landmark 2008 decision on slavery in R v Tang [2009] 2 LRC 592. That slavery covers de facto treatment as a slave, is now a well-established principle of international law, international criminal law, and is implicit in the English authority of R v K and the fact that slavery is indictable in England and Wales under the MSA. In Basfar v Wong Lords Briggs and Leggatt interpreted slavery in international law to include de facto slavery [75-76]. 

The upshot of these developments is that legally, modern slavery is slavery. That is not to require or invite qualitative comparisons between modern slavery and the barbaric crimes of the Atlantic slave trade. Rather, the constitutional prohibition in Somerset has evolved, consistently with normal principles of the evolution of the common and international law, to prohibit the modern forms of de facto slavery, servitude, FCL and human trafficking, in harmony with ECHR law and general international law.

How does the POS function?

Of course, whilst the scope of the conduct prohibited by the POS may match that under Article 4 ECHR, the nature of the rights and duties it gives rise to will not necessarily do so. In particular, the common law has no direct equivalent to modern ECHR law’s positive obligations. But as set out above, since Somerset the POS has included a right, enforceable by migrants, not to be removed from the jurisdiction against one’s will to slavery. That right is absolute, just as the right not to be subject to slavery within the jurisdiction is. 

Due to the overweening role of the HRA over the last 20 years, the courts have rarely considered or interpreted powers or measures that engage or put at risk absolute CLCR. But in Unison, where there was “a real risk that persons would effectively be prevented from having access to justice” [87], the Supreme Court Tribunal Fees order was ultra vires. As Varuhas notes, that is a test based on a form of right of access to the court which is absolute – the right against a total bar from access to court is an absolute one [596-597]. The absolute common law right to a fair trial is also breached where there is a real risk a jury has become prejudiced against a defendant. In Belmarsh No.2 a minority in the House of Lords consisting of Lords Bingham, Hoffmann and Nicholls, considered that a real risk that evidence had been obtained by torture (as opposed to a probability) was the correct threshold for exclusion of evidence by SIAC. The minority’s approach is preferable. The threshold of a real risk of breach provides the level of protection appropriate to prevent a breach which the common law will not countenance in any circumstances. In any event, the low threshold of a real risk is to be preferred in circumstances where the relevant rights breach has not yet occurred (which was not the case with torture in Belmarsh No.2; the question was one of admissibility after the fact). A real risk threshold also has the benefit of according with the threshold for the Soering test in ECHR law. That has not yet been applied in an Article 4 ECtHR case, but the structure of the unqualified rights under the ECHR is generally the same. The test in Soering has stood the test of time and been applied across Articles 2 and 3. 

The modern common law English POS therefore includes a right not to be deported to a real risk of slavery, servitude, forced or compulsory labour or human trafficking.

The risk of re-trafficking and Rwanda

There is wide recognition that modern slavery victims are at particular risk of re-trafficking. In R (TDT) v Secretary of State for the Home Department, Underhill LJ held in relation to the Article 4 protection duty at [41]:

[H]aving [been trafficked in the past] is the paradigm case of someone who is likely to be at real and immediate risk and so require protection (…) In order to decide whether a past victim is indeed no longer at real and immediate risk of being (re-) trafficked the authorities will in any event have to conduct a careful assessment of the kind for which they contended.

The Home Office recently conceded a judicial review of the SSHD’s Public Order Disqualification policy under the provisions of the Nationality and Borders Act 2022 brought in part on the basis that a re-trafficking assessment should be undertaken in every case before modern slavery protections can be removed, and updated the MSA statutory guidance to reflect that requirement [14.236].

If such a re-trafficking risk arises in the UK, which has one of the most developed anti-trafficking regimes in Europe, it must apply a fortiori in Rwanda, which as a Tier 2 Country does not meet the minimum standards in the US State Department’s annual Trafficking in Persons Report. The US concerns are reiterated in the Home Office’s Country Information Note on Rwanda and the published UNHCR Evidence. In its 2021 Evaluation Report on the UK, the Group of Experts on Action against Trafficking In Human Beings urged the UK to review its victim return and repatriation policies. The Rwanda Treaty, to be given effect by the RA, states at Article 13 only that “Rwanda shall have regard to information provided about a Relocated Individual relating to any special needs that may arise as a result of their being a victim of modern slavery or human trafficking, and shall take all necessary steps to ensure that these needs are accommodated”, and that Rwanda agrees to treat positive reasonable grounds decisions for relocated individuals as a finding they are a VOT. It contains no assurances regarding, or commitment to, a re-trafficking risk assessment, and no evidence is available as to how slavery victims’ protection needs are to be met under the Rwanda scheme. 

As matters stand, therefore, without careful assessment of their risks of re-trafficking and specific vulnerabilities as a slavery victim, in the light of the conditions on the ground in Rwanda, suspected and confirmed victims of slavery must be assumed to be at real risk of slavery. On that basis the common law constitutional POS will be engaged in every case of proposed removal. 

Interpreting s.4 in the light of the POS 

Returning to s.4 of the RA, it states that the Act does not prevent:

(a) the Secretary of State or an immigration officer from deciding … whether the Republic of Rwanda is a safe country for the person in question, based on compelling evidence relating specifically to the person’s particular individual circumstances (rather than on the grounds that the Republic of Rwanda is not a safe country in general),

…

First, read in context, where s.4(a) is engaged it will prevent removal altogether. If a decision maker finds Rwanda is not safe for an individual, then it will by definition not be a safe third country for their removal under the Immigration Acts. 

Second, subsection (2), which states s.4(1) “does not permit a decision-maker to consider any matter, claim or complaint to the extent that it relates to the issue of whether the Republic of Rwanda will or may remove or send the person in question to another State in contravention of any of its international obligations” will not apply to the POS issue, because it relates specifically to state refoulement and not to human trafficking or slavery. A person’s status as a victim of slavery is by definition something that applies to their individual particular circumstances and so falls under s.4(a).  

Third, read in the light of the common law right not to be removed to a real risk of slavery, the s.4 exceptionshould be read to apply to all suspected or confirmed modern victims of slavery, where a detailed assessment of their own specific risk of re-trafficking in Rwanda has not been carried out. That is because it must be read in a way that respects the absolute common law right not to be forcibly removed to a real risk of slavery. That is an application of a form of the principle of legality as set out by Lord Hoffmann in Simms. As Lord Browne-Wilkinson made clear in Pierson (as adopted by Lord Reed in AXA v Lord Advocate [52]), a statutory power given in general terms “cannot be construed so as to authorise the exercise of the power… in any manner which would… conflict with the basic principles of the law of the United Kingdom.” There are no words which in terms or by necessary implication exclude the POS from consideration under s.4(a). 

Fourth, read on that basis suspected and confirmed slavery cases will also meet the threshold for interim relief preventing or delaying removal, which is preserved under s.4(4) cases where claimants “face a real, imminent and foreseeable risk of serious and irreversible harm if removed…” Their removal should thus be prevented altogether.

Conclusion

There will be many legal arguments made against removal under the RA. It may well be that no directions to remove victims of slavery are ever set. It may also be that the POS could be put to other arguments against removal relating to the safety and protection needs of slavery victims. However, there is merit in analysing the scope of the POS in any event, in order to ensure that “the evolving requirements of humanity enshrined in the common law” are identified [Elgizouli, per Lord Kerr in dissent at 66]. 

            
I would like to thank Se-shauna Wheatle, Paul Scott, James Lee and Chris Buttler KC for comments on previous drafts.

James Robottom is a barrister at Matrix Chambers and a PhD candidate at King’s College London.

(Suggested citation: J. Robottom, ‘The Safety of Rwanda Act, Slavery and the Common Law’, U.K. Const. L. Blog (25th April 2024) (available at https://ukconstitutionallaw.org/)