Blog Post by Dr Sara Arapiles, Postdoctoral Fellow at Lund University; Visiting Research Fellow at the Refugee Studies Centre, University of Oxford. This post is part of the blog series on the application of refugee law in military service-related cases.

This work is supported by the Economic and Social Research Council (grant number es/p000711/ 1).

Can compulsory military service breach human rights protections against slavery, servitude or forced or compulsory labour contrary to Article 8 of the International Covenant on Civil and Political Rights (‘ICCPR’) (and/or equivalent provisions in regional human rights instruments) and, in doing so, meet the threshold of persecution under the 1951 Convention Relating to the Status of Refugees? The decision of the United Kingdom’s Upper Tribunal (Immigration and Asylum Chamber) (‘UTIAC’) in MST and Others (national service – risk categories) Eritrea CG [2016] UKUT 443 (IAC) (‘MST’) represents, to my knowledge, the first case in which an asylum court or tribunal was engaged substantively with this question. While I suggest that the factual and legal analysis in MST is imperfect, the UTIAC’s examination of the claim encompassing the protections I have identified deserves acknowledgement. Prior to this case, courts and tribunals worldwide had generally, in military service cases, focused (as some still do) instead on the risk of inhuman treatment contrary to Article 7 of the ICCPR (and/or its regional counterparts) on return for the type of punishment for desertion or draft evasion, and/or due to the conditions under which service takes place. The Country Guidance case in MST, although issued almost a decade ago, still stands as an authoritative factual conclusion in the UK regarding Eritrean asylum claims and military service. The critical discussion that now follows is important for this reason, but also because of the increased significance of military service today due to the escalation in armed conflicts globally and the reemergence of compulsory military service, including conscription, in various states.

In MST, the UTIAC assessed whether Eritrea’s Military/National Service Programme (‘MNSP’) gave rise to a real risk of exposure to treatment contrary to Article 4 of the European Convention on Human Rights (‘ECHR’) and/or Article 8 of the ICCPR – the prohibition of slavery, servitude, and forced or compulsory labour – on return, and constituted persecution under the 1951 Refugee Convention.  Article 4 of the ECHR provides as follows:

‘Article 4 Prohibition of slavery and forced labour

1. No one shall be held in slavery or servitude.

2. No one shall be required to perform forced or compulsory labour.

3. For the purpose of this Article the term “forced or compulsory labour” shall not include:

(a) any work required to be done in the ordinary course of detention imposed according to the provisions of Article 5 of this Convention or during conditional release from such detention;

(b) any service of a military character or, in case of conscientious objectors in countries where they are recognised, service exacted instead of compulsory military service;

(c) any service exacted in case of an emergency or calamity threatening the life or well-being of the community;

(d) any work or service which forms part of normal civic obligations.’

UNHCR, which intervened as a third party, invited the UTIAC to find specifically that a return to the MNSP would amount to a breach of the prohibition of servitude (Article 4(1)) or of the prohibition of forced or compulsory labour (Article 4(2)) not falling within the Article 4(3) exceptions. UNHCR however, remained silent on slavery within Article 4(1). Most pronouncements and literature on the MNSP also do. Although some commentators have described service in the MNSP as a form of ‘modern slavery’ or a ‘slavery-like’ practice, such terms have generally been used rhetorically, not factually and/or legally to mean slavery stricto sensu. Such terminology creates, in my view, an unhelpful “old/modern” dichotomy, contributing to the erroneous belief that slavery is confined to chattel slavery and, therefore, belongs only to the past. This may be the reason why UNHCR did not make any submission to the UTIAC about slavery in reference to the MNSP.

Despite UNHCR’s omission, the UTIAC analysed in detail whether the MNSP amounted to any of the three proscribed conducts covered in Article 4 of the ECHR, including slavery. I address these in turn below.

Forced or Compulsory Labour (Article 4(2)(3) of the ECHR)

In MST the UTIAC ultimately held that the MNSP fell within the meaning of forced or compulsory labour in Article 4(2) of the ECHR and outside the scope of the exclusion criteria in Article 4(3). In the analysis preceding this conclusion, the UTIAC treated the International Labour Organisation (‘ILO’) framework as a relevant source of interpretation of Articles 4(2) and 4(3) of the ECHR. The UTIAC also drew on the ILO’s interpretation of forced labour within the MNSP.

Guided by the approach taken by the European Court of Human Rights (‘ECtHR’) in prior jurisprudence concerning Article 4 of the ECHR, the UTIAC in MST looked at Article 2(1) of the 1930 ILO Forced Labour Convention. This defines forced or compulsory labour as: ‘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’. The UTIAC also referred to the decision in Van Der Mussele v Belgium, observing that the ECtHR had ‘approved the European Commission of Human Rights’ position  [in Iversen v. Norway] that … two [emphasis added] cumulative conditions have to be satisfied’ for a phenomenon to constitute forced or compulsory labour under Article 4(2) of the ECHR: ‘not only must the labour be performed by the person against his or her will, but either the obligation to carry it out must be “unjust” or “oppressive” or its performance must constitute “an avoidable hardship”’. However, this observation is partially wrong. In Van Der Mussele v Belgium, while the ECtHR endorsed the first condition in Iversen v. Norway, it distanced itself from the second factual element. For the ECtHR, ‘unjust’, ‘oppressive’ or ‘avoidable hardship’ are not sine qua non features of forced or compulsory labour (a verdict that has been recalled and endorsed in the jurisprudence of the International Criminal Tribunal for the Former Yugoslavia and which is consonant with the definition of forced or compulsory labour in the 1930 ILO Forced Labour Convention).  

In MST the UTIAC then moved to consider the prescribed exceptions or exclusions to forced or compulsory labour listed in Article 4(3) of the ECHR (which mirrors those in Article 2(2) of the 1930 ILO Forced Labour Convention). In doing so, it held that the extended obligations imposed on Eritrean conscripts ranging from ‘ordinary public works’ to work in the ‘private sector’ go far beyond work or service that is of a ‘military character’. As widely documented, besides military duties, Eritrean conscripts are forced to work on private construction sites, farms, mining operations, infrastructure projects, enterprises owned and operated by the government, and for the government’s development programmes. In addition, the government “lends” conscripts to foreign companies operating in Eritrea. Other conscripts also perform civilian duties in the administration, ministries, schools, hospitals, and judiciary. These duties extend indefinitely. Proclamation No. 82/1995 (which regulates the MNSP) establishes that the service demanded from Eritrean conscripts is used for the purposes of, among other things, national reconstruction and economic development. It also states that military service includes ‘active military service and development tasks [emphasis added] in military forces’.

While the exception to forced or compulsory labour under both the ECHR and ICCPR refers (only) to any service of a ‘military character’, the 1930 ILO Forced Labour Convention speaks of ‘purely [emphasis added] military character’. In the words of ILO, the term ‘purely’ aims specifically at preventing the requisitioning of conscripts for the performance of public works (a claim endorsed by UNHCR Protection Guidelines related to Military Service). In this regard, the UTIAC in MST further remarked that the need for compulsory military service to be of a ‘purely military character’ has its corollary in Article 1(b) of the 1957 ILO Abolition of Forced Labour Convention, which prohibits the use of forced or compulsory labour ‘for the purposes of economic development’. This Convention also prohibits the use of forced labour, inter alia, as a means of ‘political coercion or education’ and ‘labour discipline’. Besides economic development, Proclamation No. 82/1995 also establishes that the objectives of the MNSP are the elimination of ‘sub-national feelings’ and the creation of a ‘new generation characterized by love of work, discipline …’ which should be understood as also prohibited by international labour standards.

Servitude (Article 4(1) of the ECHR)

In the MST case, the UTIAC adhered to a passage in Siliadin v France (which in turn relates to a passage from the Commission report in the earlier Van Droogenbroeck v Belgium case) which states that servitude implies ‘in addition to the obligation to perform certain services for others … the obligation for the “serf” to live on another person’s property and the impossibility of altering his condition’. MST further quoted the following excerpt from the ECtHR’s judgment in C.N. and V v France: ‘the fundamental distinguishing feature between servitude and forced or compulsory labour … lies in the victim’s feeling that their condition is permanent and that the situation is unlikely to change’. In its assessment, the UTIAC, however, disregarded the latter case (probably because it cited the above excerpt from the ECtHR’s Guide on Article 4 of the ECHR and not from the case itself), and went to hold that the MNSP did not constitute ‘anything comparable to the paradigm identified in Siliadin’. The UTIAC continued: (a) Eritrean conscripts, even those ‘who are required to perform lengthy [MNSP] cannot sensibly be described as being compelled to live permanently on government property’;and (b) while ‘possibilities for exemption or de facto demobilisation are limited, it cannot be said that there is an impossibility to alter one’s condition’.

The first point, on the requirement of ‘being compelled to live permanently on government property’, is clearly wrong. In Siliadin – which the UTIAC claims to use as the definitive benchmark – the ECtHR did not refer to a ‘permanent’ obligation to stay at a location imposed by someone else but rather to the impossibility of leaving of one’s own will. The word ‘permanent’ is to be found nowhere in the Siliadin decision; it only appears in the latter case of C.N. and V (to which I return shortly). In addition, in Siliadin (and also in C.N. and V), the ECtHR removed the obligation to live at someone else’s property in the Commission’s report in Van Droogenbroeck from its interpretation of servitude. The Commission, for its part, had taken such a requirement from the definition of serfdom in the 1956 Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery. In Siliadin, however, the ECtHR held that, for the purposes of the ECHR, ‘“servitude” means an obligation to provide one’s services that is imposed by the use of coercion, and is to be linked with the concept of “slavery”’. Drawing on this, the Strasbourg Court in C.N. and V further observed that ‘servitude corresponds to a special type of forced or compulsory labour or, in other words, “aggravated” forced or compulsory labour’. The ECtHR therefore does not consider the obligation to reside on someone else’s property to meet the threshold of servitude under Article 4(1) of the ECHR per se.

The UTIAC’s assessment of the second point, i.e., the impossibility of altering one’s condition, is also flawed. In C.N. and V., building on the Siliadin case, the ECtHR construed it as ‘the victim’s feeling [emphasis added]’– brought about objective criteria – of their condition/situation being ‘permanent and … unlikely to change’. In measuring whether a situation classifies as servitude or slavery, the relevance does not lie, at least exclusively, on the (permanent) duration of the relationship between the serf/slave and the serf-/slaveholder (which in C.N. and V. was of four years), but on the quality of it. In its own findings in MST, the UTIAC explicitly stated that discharge or release from the MNSP is ‘arbitrary’ and ‘limited’ as, in its own words, ‘it is at the whim of a commander or employer’. UTIAC’s findings also point to the MNSP being ‘arbitrary and indefinite’ to the extent that the period that a person starting the MNSP will be expected to complete ‘is not known’. Throughout the judgment, the UTIAC also paid attention to other relevant factors, such as the imposed impossibility of the conscript’s pursuing their own occupations and professions, and the statutory penalties and extrajudicial forms of punishment attached to disciplinary offences, desertion, draft evasion, and exit from the country. It is reasonable to argue that all these elements combined would make Eritrean conscripts feel that their condition is arbitrarily defined and ‘unlikely to change’ and, therefore, ‘permanent’ in their eyes. The UTIAC, however, failed to properly engage with its own findings in its legal analysis of servitude (and slavery) and, instead, placed special emphasis on a lifelong/permanent situation, which led the Chamber to find that the MNSP falls outside the meaning of servitude in Article 4(1) of the ECHR.

Furthermore, MST relied on sporadic evidence indicating that release or discharge from the MNSP may be possible, for instance, by paying bribes or corruption to support its assessment. This argument should have led the UTIAC to a positive finding on servitude instead. Indeed, in C.N. and V., the ECtHR found that the concerned victim was subjected to servitude as, inter alia, she ‘was convinced [emphasis added] … that she could not free herself from their hold without placing herself in an illegal [emphasis added] situation’ – it did not have to be proven that departure was totally impossible because of physical force. In Eritrea, conscripts can only free themselves from their slaveholder by committing the crime of exiting the country or, in very rare cases, by corruption. Moreover, having to assert one’s own freedom with a ransom may constitute – at the very last – a situation of ‘debt bondage’, which is one of the types of servitude covered in the 1956 Supplementary Convention.

In reviewing the UTIAC’s understanding of servitude in MST, it has become clear that the overlap between Siliadin and C.N. and V. made by the UTIAC is not faithful to the Strasbourg Court’s interpretation of the concept of servitude in Article 4(1).

Slavery (Article 4(1) of the ECHR)

Having looked at Strasbourg Court’s judgment in Rantsev v Cyprus and Russia, the UTIAC held that the obligation to perform the MNSP cannot be described as ‘the exercise by the Eritrean state of a genuine right of legal [emphasis added] ownership reducing those called up to the status of an “object”’. It added: ‘Eritrean law does not create such a legal ownership’. This led the UTIAC to make a negative finding on slavery. However, in Rantsev, the ECtHR did not speak of a ‘genuine right of legal ownership’, but instead of a ‘right of ownership’. The understanding that slavery under Article 4 of the ECHR requires a ‘genuine right of legal ownership’ draws instead from Siliadin, the first case in which the ECtHR addressed a claim based on Article 4(1). The UTIAC thus overlooked that, in later jurisprudence, including Rantsev, the ECtHR had removed the requirement of ‘legal’ ownership. Moreover, the UTIAC’s interpretation of slavery does not correspond to the wording of the definition of slavery in Article 1(1) of the 1926 Slavery Convention (which is replicated in Article 7(a) of the 1956 Supplementary Convention and in Article 7(2)(c) of the 1988 Rome Statute of the International Criminal Court) which instead refers to the exercise of ‘any or all of the powers attaching to [emphasis added] the right of ownership’ over a person. Due to its almost universal acceptance, the 1926 definition has reached customary international law status and cannot simply be ignored. ‘Powers attaching to’ reads in the French version of the Convention as ‘les attributs’.  As Allain explains, the slavery definition does not speak of legal ownership, ‘but of exercising the attributes of the right of ownership without necessarily exercising a legal right of ownership’. 

The exercise of powers attaching to the right of ownership translates into the exercise of a substantial degree of control over the victim which in turn enables the slaveholder to, inter alia, enjoy, exploit, dispose of, profit from, transfer, lend, and/or use the slave at their whim and for whatever purpose – it can also be in furtherance of war efforts. For instance, the Special Court for Sierra Leone made positive findings on slavery with regard to the civilians forced to undergo military training – these were abducted ‘to achieve the [Armed Forces Revolutionary Council and the Revolutionary United Front] RUF/AFRC’s military and political goals, specifically in order to support and sustain the RUF/AFRC and enhance its military capacity and operations’. Importantly, slaves may also be required to do nothing. As the Office of the Prosecutor of the International Criminal Court notes: slavery does ‘[not] require that the person who is enslaved “do” anything’.

Courts and tribunals that have dealt with slavery cases, not only within the international criminal law framework, but also the Inter-American Court of Human Rights, undertake a ‘multi-factor analysis’ for the purposes of identifying a case of slavery. Factors that may indicate the existence of slavery include: (i) restriction or control of an individual’s autonomy; (ii) restriction or control of someone’s movement, including measures taken to prevent or deter escape; (iii) restriction or control of  freedom of choice; (iv) control of physical environment; (v) psychological control or oppression; (vi) force, threat of force, coercion, fear of violence, deception or false promises; (vii) duration; (viii) assertion of exclusivity; (ix) subjection to cruel treatment and abuse; (x) control of sexuality; (xi) forced labour; (xii) the abuse of power; (xiii) the victim’s position of vulnerability; (xiv) detention or captivity; and (xv) the socioeconomic conditions in which the situation takes place. While these factors are indicative of slavery, the presence of all is not queried to satisfy the definition of slavery. For instance, a person may be subjected to slavery for purposes other than forced labour, or ‘slavery may exist even without torture’. Only the exercise of control over a person in a way that restricts or deprives a person of their individual liberty and/or autonomy, or that results in the person’s loss of own will is ‘the sine qua non of slavery’. This type of control while can be exercised through physical constraints, it can also be maintained through threats of force or death, coercion, and even deception or fraud. As the International Criminal Court put it, it covers situations in which the victim is ‘unable to leave as they would have nowhere else to go and fear for their lives’. It can also manifest as omissions, for instance, through ‘denials of access to rights or degradation of human dignity and humane existence’.

By failing to engage normatively with the definition of slavery, the UTIAC in MST sidelined relevant factors that classify the MNSP as slavery. The UTIAC made findings on forced labour, lack of freedom of choice, measures taken to prevent escape, arbitrary detention, disproportionate punishment, the indefinite duration of the MNSP, and the impossibility of being released from the MNSP by the will of the people subject to it. All of these are factual indicators of slavery as defined in the relevant instruments. Furthermore, and most importantly, the UTIAC also emphasised that the available evidence strongly suggests that the policy imperatives of the Eritrean government are driven ‘by domestic concerns about the maintenance of control and regulation of their own population’ further remarking that the MNSP ‘is a way of controlling the population’. In considering this factual element and the aforementioned factors in isolation from each other, the UTIAC overlooked their significance – as indicia of slavery – to reach incorrect interpretations.

Forced or compulsory labour as persecutory harm

It shall be recalled that the UTIAC in MST made positive findings of forced labour in the MNSP. The UTIAC further considered that to the extent that the MNSP breaches the prohibition of forced or compulsory labour in Article 4(2) of the ECHR, it is also likely to constitute a breach of Article 3 of the ECHR (inhuman treatment). The UTIAC did not deem it necessary to consider Article 3 separately but instead assimilated the right not to be subjected to forced or compulsory labour to the right not to be subjected to inhuman treatment, finding a factual overlap between both provisions. Whether this assessment is due to having wrongly attributed the elements of ‘oppressive’ or ‘avoidable hardship’ to the definition of forced or compulsory labour is not clear. Regardless, MST concluded that, generally, any person having to serve in the MNSP on return will face a real risk of persecution or serious harm or ill-treatment contrary to Article 4(2) or Article 3 of the ECHR, even if such individual ‘may avoid punishment in the form of detention and ill-treatment’. It thus found that forced participation in the MNSP constitutes persecutory harm in its own right.

The question that now arises is: persecution on what Convention ground? In MST the UTIAC found service in the MNSP to constitute persecution on the Convention ground of imputed political opinion, because, among other things, the Eritrean government regards the MNSP as a ‘patriotic’ duty. Additionally, while the UTIAC did not expressly refer to the ground of ‘membership of a particular social group’, it specified the various groups that, on return, face a real risk of persecution, which in essence include any individual of or approaching draft age who will be perceived on forcible return as draft evaders or deserters, and generally any person who face having to resume or commence the MNSP.

Concluding remarks

Compulsory military service exacted ‘under the menace of any penalty’ that goes beyond purely military service – for instance, for being used for the purposes of economic development and/or other public and private works – would constitute forced or compulsory labour falling outside the permitted exceptions or exclusions in both international human rights law and the ILO Forced Labour Conventions. If other factual elements are present, it can also constitute servitude or slavery. What is more, it can constitute the three proscribed conducts concurrently.

As illustrated above, the UTIAC’s interpretation of slavery and servitude (and some elements of the interpretation of forced or compulsory labour) in MST simply does not hold, and errs in law. The UTIAC laid down an artificially elevated and incorrect threshold to satisfy Article 4(1) of the ECHR (the right not to be subjected to slavery or servitude), drawing – wrongly – on features such as legal slavery (i.e., ownership of a person prescribed by law) and permanence (i.e., a lifelong situation). MST also narrowed down the material scope of Article 4(2) of the ECHR to require a situation of forced or compulsory labour to be ‘unjust’, ‘oppressive’, or involve ‘avoidable hardship’. Despite these shortcomings, courts in Switzerland and Germany have relied on the decision in MST at face value. In doing so, these courts relied exclusively on ‘legal ownership’, ‘permanence’, and ‘hardship’ to make negative findings on slavery and servitude, and some also on forced labour, unduly denying international protection for Eritrean applicants in some cases. Courts in other European jurisdictions have simply sidelined factors that are indicative of slavery in the MNSP. The absence of a clear and correct understanding is exacerbated by the broader paucity of refugee law scholarship and UNHCR pronouncements on these proscribed conducts.

Finally, the widespread use of the term ‘modern slavery’ leads to the flawed understanding that chattel slavery, i.e., the legal ownership of a human being, is the definitive benchmark. This terminology promotes the erroneous belief that slavery only belongs to the past, and it is against this background that a misleading picture of the phenomenon has developed. This contributes to the neglect of the normative relevance of the international law definition of slavery in the context of refugee law. This (neglected) definition is in fact the one that judges should work with to identify slavery cases, notwithstanding the recent shift in terminology whereby modern slavery is equated with all sorts of forms of human exploitation. This also applies in the UK. In the 2015 UK Modern Slavery Act, ‘modern slavery’ is an overarching term that encompasses slavery, servitude, forced or compulsory labour and human trafficking, while in the UK Nationality and Borders Act (2022) ‘modern slavery’ only covers slavery and human trafficking. UK legislation thus is not consistent in the way it uses this umbrella term, and therefore attention shall be paid to each of the proscribed conducts falling under the term separately. All these phenomena, while often simplistically termed ‘modern slavery’, remain conceptually distinct in law, and this requires recognition both generally, and specifically in the interpretation of the refugee definition.

---

The views expressed in this article belong to the author/s and do not necessarily reflect those of the Refugee Law Initiative. We welcome comments and contributions to this blog – please comment below and see here for contribution guidelines.