In November 2005 the House of Lords (HoL) delivered a landmark judgment (R v SSHD ex parte Adam, Limbuela and Tesema, UKHL. Hereafter Adam, UKHL), ending a ten-year period of legislation and contestation regarding the removal of welfare support from asylum seekers who do not claim on entry into national territory (late claimers). This history began with the Conservative government's attempt to remove benefits from late claimers by changes in Income Support Regulations (Department of Social Security, 1996) and ended with an HoL judgment on the application of Labour's 2002 (Nationality Asylum and Immigration Act (Home Office, 2002), yielding a sequence in which ‘each step taken by Parliament has led to litigation’ (R v SSHD on application of T, para. 6. Hereafter T). Of the 14 judgments in this case history, 11 went against the Government.
One interesting question is what was at stake in these judgments, and why so many cases were necessary to bring the sequence to a final conclusion. The policy at issue stems from government concern over asylum and immigration, and attempts at deterrence which collide with the assertion of fundamental rights. Thus, challenges to its implementation implicitly raise questions about the boundary of our social and moral community, and highlight the role of the judiciary in determining the legitimacy of contentious policy measures. The case history spans a period of transition in Britain from the general principles of the Common Law to implementation of the Human Rights Act (HRA), the latter requiring public authorities to act compatibly with the European Convention on Human Rights (ECHR). Yet indeterminacy is to the fore in such a developing area of law (Dworkin, Reference Dworkin1977), allowing considerable scope for judicial interpretation.
The following account of these cases documents an incremental process in which the application and interpretation of basic values and principles plays a vital role. The starting point for analysis is therefore the moment of judgment in each case: what issues the decision turns upon, what general principles the judges draw on and the impact of such principles on the outcome. These questions can then be considered alongside more overt policy issues such as the force of collective interest as against individual rights, the way judges differ in their interpretation of these matters and how these differences are resolved across a series of cases. Beyond these questions, we find a more elusive terrain of further interest to policy analysts: judgment as political dialogue, which can be explored by close attention to both the rationale and delivery of each judgment. Legomsky's (Reference Legomsky1987) account of the judicial treatment of immigration law notes the significance of ‘extra-legal’ factors in shaping a judgment, and he includes here the force of political concerns. Accordingly, we can ask what legal and political strategies judges deploy in the formulation and delivery of their judgment, what political messages the judgments contain and what impact they have on the application of policy.
Such questions open up the idea of judgment as part of a dynamic process that shapes our conception of social and political life, and establishes parameters of acceptability. Analysis of these cases is therefore of great relevance to social policy, especially given the high success rate of the challenges. The judgments allow us to examine how far constraints may be placed on government action by virtue of recourse to general principles and fundamental rights, and they have powerful implications for the deterrent approach to asylum which has been a key aspect of policy over the last decade (Stevens, Reference Stevens2004; Webber. Reference Webber2004). They also permit an assessment of the role of the HRA, as compared with common law principles, in challenging the implementation of such policy. Conversely, we can consider how far the judgments are constrained by government objectives, through an examination of the political concerns at issue and their handling by the judiciary. What is revealed is an insight into the role of judgment as part of a political exchange both between judges and with the government (cf. Fredman, Reference Fredman2000) over permissible means of deterrence.
The context for the cases under consideration is an attempt by respective governments to harness the benefits system as a means to deter arrival, each instance of the withdrawal of support coinciding with a significant rise in asylum numbers (to 43,925 in 1996, and to 84,135 in 2002). In the case of both Conservative and New Labour policy in this area, the justification for withdrawal was a claim to combat abuse, despite critical comment on the policy's failure to discriminate between genuine and non-genuine claimants.Footnote 1 It has been shown that late claims were more likely to succeed than applications on arrival,Footnote 2 and that the measure was based on misguided assumptions about how asylum seekers choose their country of destination, benefits playing a lesser role than that of traffickers and the availability of informal supports (Bloch and Schuster, Reference Bloch and Schuster2002; cf Juss, Reference Juss1997, on ‘cultural jurisprudence’). The policy has been one aspect of a broader attempt to deter and control certain types of migration (Stevens, Reference Stevens2004), linked under New Labour to an emphasis on targets for the reduction and removal of asylum seekers (Düvell and Jordan, Reference Düvell and Jordan2003). The erosion of welfare support as part of this process has been documented by Cohen et al. (Reference Cohen, Humphries and Mynott2002) and Sales (Reference Sales2002), while Morris (Reference Morris2007) has addressed migration and asylum policy in terms of a system of civic stratification, designed to attract or repel certain categories of migrant.
Judicial decision-making in the field of immigration and asylum has traditionally been driven by deference to policy priorities (Stevens, Reference Stevens2004), and shaped to a degree by conservatism and constraint (Legomsky, Reference Legomsky1987). However, many writers (Stevens, Reference Stevens2002; Woodhouse, Reference Woodhouse1998; Woolf, Reference Woolf1998) see a recent growth of Judicial Review as marking increased judicial involvement in placing government policy under challenge. Woodhouse, for example, cites a number of cases reading positive rights to social support from protections such as the right to life or to seek asylum, and she predicts an increase in judicial intervention to defend rights historically associated with the welfare state. The present history is therefore of particular interest, in dealing with welfare issues which are indirectly linked to the immigration aspect of asylum policy. Early cases were confined to interpretations of domestic law, albeit read against the Convention on the Status of Refugees (CSR) and the Common Law. Later cases post-dated the HRA and invoked an absolute human right, requiring the highest degree of scrutiny and offering little scope for qualification in terms of the national interest. Both before and after the passing of the HRA, key judgments ruled against the application and effect of government policy, and below we examine their unfolding logic and policy implications. Other articles from this project will examine civil society involvement in bringing the challenges (Morris, Reference Morris2009b), the implications of the cases for debate on cosmopolitanism (Morris, Reference Morris2009a) and the wider social impact of the judgments (Morris, Reference Morris2009c).
Before the HRA
Though early cases have been documented elsewhere (Feria, Reference Feria1996; Gillespie, Reference Gillespie1996; Scoular, Reference Scoular1997), they provide an interesting backdrop to the later HRA cases and are briefly summarised below, with particular attention to the political thrust of the judgments and the role played by general principles in the interpretation of domestic law.
The Law of Humanity
The first case arose from a challenge to the 1996 Income Support Regulations under the Conservative government, which abolished benefits for asylum seekers who did not claim asylum immediately on arrival, or were appealing a first decision. The challenge was initiated by the Joint Council for the Welfare of Immigrants (R v SSSS ex parte JCWI, QBD), alleging inconsistency with the 1951 Convention on the Status of Refugees (CSR) and the 1993 Asylum and Immigration Appeals Act (Home Office, 1993). The High Court (HC) found no direct interference with rights conferred by the 1993 Act (Scoular, Reference Scoular1997), and that the removal of benefits did not amount to deportation, expulsion or refoulement.Footnote 3 Nevertheless, the Court of Appeal (CA) reversed the decision (R v SSSS ex parte JCWI [1966] 4 All ER 385. Hereafter JCWI, CA). Simon Brown LJ noted the questionable logic of the policy and accepted an argument of conflict with asylum seekers’ right to status determination, contained in the CSR and confirmed in the 1993 Act. In his view, enforced destitution of asylum seekers did amount to constructive refoulement, and could place the UK in violation of the CSR.
The ostensible core of the judgment was that statutory rights were not to be cut down by subordinate legislation, but the decision was much more broadly conceived, arguing that Parliament made a full commitment to the CSR in the 1993 Act, granting fuller rights than asylum seekers had previously enjoyed, some of which would be rendered nugatory by the Regulations. Simon Brown LJ therefore identified an expansionary logic at work, which would be undermined by the Regulations, and ruled that ‘Primary legislation alone could achieve that sorry state of affairs’ (JCWI, CA, p. 13). Criticism of the government went further, in describing the ‘uncompromisingly draconian’ (JCWI, CA, p. 13) effect of the Regulations which ‘necessarily contemplate a life so destitute that . . . no civilised nation can tolerate it’ (JCWI, CA, p. 12). Simon Brown LJ also made reference to an earlier case (R v Eastbourne, 1803), which upheld an obligation to maintain poor foreigners by reference to the ‘Law of Humanity’ (JCWI, CA, p. 12), and moved beyond the core of the legal decision to state:
So basic are the human rights at issue that it cannot be necessary to resort to the ECHR to take note of their violation. (JCWI, CA, p. 12)
In effect, the Law of Humanity gave moral weight to the decision, underlining the unacceptable nature of the policy, while Simon Brown LJ's comment on the ECHR issued an implied threat or warning of a bigger debate in waiting, to be played out in later cases. In contrast, the dissenting opinion from Neill LJ stressed the established treatment of persons from abroad as special cases, the primary purpose of the Regulations given rising public costs and the need to strike a balance between collective interests and the rights of asylum seekers. These issues drive much of the ensuing history, and Neill LJ's position is implicit in later attempts to re-establish the policy. Since this case turned on the legality of secondary legislation, it remained open to the government to confirm the policy in primary legislation, which in turn sparked further challenge.
The National Assistance Act
The 1996 Asylum and Immigration Act (Home Office, 1996) enshrined key aspects of the discredited Regulations in primary legislation (see Gillespie, Reference Gillespie1996) and so overcame the negative ruling. The next case (R v Westminster Council ex parte M [1977] 1 CCLR 69. Hereafter M) turned upon S21(1)(a) of the 1948 National Assistance Act (NAA) (Department of Health, 1948), granting residential care to those ‘in need of care and attention’, now put to a use in apparent conflict with the purpose of the 1996 Act. Collins J, in his HC judgment, recognised the intention of Parliament to reverse the JCWI decision (M, p. 13), but in failing to designate exclusion from the NAA they had left open the question of Local Authority liability. Since Parliament ‘could have made (this) clear beyond doubt’ (M, p. 15), the interpretive task was to uncover their intentions in the NAA, which was deemed a source of provision for those unable to fend for themselves. Resource constraints were ruled out of the picture and Collins J cited the JCWI case on the dilemma confronting asylum seekers without benefit and on asylum rights rendered valueless.
As in the Court of Appeal (CA) judgment on JCWI, once the legal core of argument was established there were a number of wider-ranging references, notably to the Right to Life and the Law of Humanity (M, p. 27). Emphasising that the judgment did not frustrate but implemented the will of Parliament as expressed by the NAA, it was declared impossible to believe that Parliament intended lawfully present asylum seekers to be left destitute and at risk of illness and death. If the intent was to deny all assistance, then:
(Parliament) must say so in terms. If it did, it would almost certainly put itself in breach of the ECHR and of the Geneva Convention. (M, p. 28)
The same judicial technique was in play as in the JCWI CA judgment, by reference to a broader argument awaiting a public airing. However, the subsequent CA judgment on M (R v Westminster Council ex parte M [1997] 1 CCLR 85, CA. Hereafter M, CA) while arriving at the same conclusion, was cast in narrower terms.
In this CA judgment, delivered by Woolf LJ, the NAA was construed as ‘an act which is always speaking’, requiring a reading which addressed contemporary circumstances. Late claimers thus qualified for support under the NAA as a result of their particular predicament: lacking food and accommodation, without knowledge of the language or country and under stress in pursuing an asylum claim. However, the judgment rested purely on a statutory interpretation of the NAA, making explicit the fact that provision could not be claimed by those simply short of money (thereby excluding undocumented migrants). It made no reference to fundamental Human Rights, the Law of Humanity or the CSR, and its impact was explicitly linked to the practical circumstances confronting asylum seekers. The policy message was more limited as a result, but nevertheless made an explicit connection between the circumstances of asylum seekers and their need for basic maintenance. It may therefore seem surprising that the denial of benefit for late claimers should resurface some five years later, under a New Labour government. However, nothing in the judgments above prohibited the removal of asylum seekers from eligibility under the NAA, or the introduction of further deterrent measures.
Rights Brought Home
By the time of the next case, the 1998 HRA had been implemented by New Labour, giving effect to the ECHR in British law. Late claimers for asylum had been explicitly excluded from the NAA, but granted maintenance and accommodation under the 1999 Immigration and Asylum Act (Home Office, 1999), via a system of dispersal administered by the National Asylum Support System (NASS) (Thomas, Reference Thomas2003). However, the 2002 Nationality Immigration and Asylum Act (Home Office, 2002 carried strong echoes of the 1996 exclusions. It denied the Secretary of State power to support asylum seekers who did not claim ‘as soon as reasonably practicable’ (S55(1)), but incorporated human rights principles by permitting support where necessary to ‘avoid a breach of Convention rights’ (S55(5)).
A number of rights were potentially at play in relation to S55(5) (Billings and Edwards, Reference Billings and Edwards2004), but article 3 was the most debated with respect to whether enforced destitution amounted to inhuman and degrading treatment (IDT) and was therefore impermissible. The Joint Committee on Human Rights (Joint Committee on Human Rights, 2002) had already noted:
It is difficult to envisage a case where a person could be destitute without there being a threat of a violation of Articles 3 and/or 8 of the ECHR. (para. 15)
Eligibility for support under NASS was anyway confined to those destitute or likely to become so within 14 days; defined in the 1999 Act (S95(3)) as the absence of adequate accommodation or the inability to meet essential living needs. S55(5) of the Nationality Immigration and Asylum Act (2002) implicitly assumed a gap between NASS destitution and any definition which might qualify for article 3 protection, without which the section would be meaningless. The effectiveness of the policy therefore hung upon the court's interpretation of IDT, and its application in the cases at issue, attaching powerful policy implications to the ensuing judgments.
Interpreting Section 55
An assault on democracy?
In the first case to challenge disqualification from support under S55 (R v SSHD on application of Q, EWHC. Hereafter Q, EWHC), Collins J drew attention to a lack of Parliamentary scrutiny (para. 5), registered JCHR concern (para. 8) and queried the overall logic of the measure (para. 14). However, the first element of the judgment was procedural. It found the interpretation of ‘as soon as reasonably practicable’ to be inadequate in neglecting the subjective element deriving from the power of agents over their charges and the possible trauma of flight. The policy impact of this finding is underlined by the Home Office (HO) argument that this interpretation would rob S55 of force and therefore undermine government policy. This argument was dismissed.
The decision on S55(5) carried greater import. While Collins J recognised that ‘Parliament (had) now in terms removed the Law of Humanity’ (Q, EWHC, para. 62),Footnote 4 there was nevertheless a possibility that the HRA could have a constraining effect. Indeed, the idea that some Convention right was at issue was implicit in S55(5) itself, and the judgment thus turned to article 3. The HO argument against entitlement to support under article 3 looked to a European Court of Human Rights ruling (O'Rourke v UK, ECtHR. Hereafter O'Rourke), which found that a British citizen living on the street in a state of poor health did not attain the ‘requisite level of severity’ (Q, EWHC, para 62). This case was subject to scrutiny in later judgments, but Collins J looked instead to the ‘real risk’ test of article 3 applied by the ECtHR in cases of threatened deportation. Satisfied that real risk of a breach existed for the present claimants, he ruled that provision could be anticipatory, and that no attention had been given to human rights issues by NASS officials. The implementation of S55 was therefore fundamentally flawed.
Collins J closed by recognising that his decision would weaken the anticipated effect of S55(1), but with a reminder of the intentions of Parliament in including S55(5), and a glancing reference to possible breach of the CSR (Q, EWHC, para. 74). In fact, once more the strategic delivery of the judgment ranged more widely than the detail of the argument, in its implied criticism of the rationale of the policy and in citing general principles which supplement the specifics of the decision. These include not only the rights enshrined in the ECHR, but also the Universal Declaration of Human Rights (UDHR) and ‘the constitutions of many civilised countries’ (para. 87). Collins J also recalled the Law of Humanity, despite its acknowledged removal:
It would be surprising if the standards of the ECHR were below those believed 200 years ago to be applicable to the law of humanity. (Q, EWHC, para. 72)
So again, we find an implicit assumption of an expansionary logic to rights, as in the JCWI CA case.
Heated responses from press and Parliament (Bradley, Reference Bradley2003) supply the context for the subsequent CA decision, and David Blunkett's outrage was widely cited:
Frankly, I'm personally fed up with having to deal with a situation where Parliament debates issues and the judges then overturn them.Footnote 5
He claimed that as a result, democracy itself was under threatFootnote 6 and thus seemingly challenged the role of the courts in interpreting the HRA in a manner which potentially undermined government policy.
Resolving a conundrum
The CA judgment on the HO appeal in Q described the case as presenting a conundrum: ‘Can the Secretary of State refuse support to the destitute without thereby subjecting them to inhuman treatment?’ (R v SSHD ex parte Q, EWCA, para 3. Hereafter Q, EWCA). Despite the praise for Collins’ approach, his ‘real risk’ solution was overruled, the parliamentary evidence was held to be inadmissible, his reasoning was termed opaque and there were important differences of detail in the treatment of S55(5). On S55(1) there was broad agreement; Philips LJ found the decision-making procedure was neither fair nor fairly operated, and rejected the HO argument that S55 would become ‘a dead letter’ (Q, EWCA, para. 39). However, assurances were offered:
When appropriate procedures are in place . . . it will be very difficult . . . to discharge the burden of proving that it was not reasonably practicable. . .to claim asylum any earlier. (para. 42)
Thus far, the judgment reads as an endorsement of the policy, but Philips LJ then turned to the applicability of article 3 – the ‘cutting edge of human rights jurisprudence’ (para. 52). The issue was whether failure to provide support could breach the negative obligation to refrain from IDT, or whether article 3 could only in extreme circumstances impose a positive obligation. The decision turned on the meaning of ‘treatment’, which implies something more than passivity, and Phillips LJ found that the denial of support under S55(1), in combination with a prohibition on employment, did indeed qualify (Q, EWCA, para. 57). This gave the CA judgment a huge additional significance, linking withdrawal of support to the breach of a negative obligation and confirming the absolute nature of article 3 as a right that cannot be qualified according to circumstance. It placed the responsibility for creating destitution firmly with the government and made arguments of balance between policy concerns and individual rights groundless.
For a definition of inhuman and degrading, the judgment turned to Pretty, an ECtHR case:
Ill-treatment that attains a minimum level of suffering and involves actual bodily injury or intense physical or mental suffering. Where treatment humiliates or debases an individual showing lack of respect for, or diminishing, his or her human dignity or arouses feeling of fear, anguish or inferiority capable of breaking an individual's moral and physical resistance. (quoted in Q, EWCA, para 60)
As in the HC, weight was placed on the faulty procedure for applying S55(5), but having supplied the meaning of IDT, it remained to furnish a test and ‘real risk’ was deemed inappropriate. It was argued that, in order to meet the requirements of article 3, the asylum seeker must have descended to such a state of vulnerability that there was high risk of an almost immediate breach (Q, EWCA, para. 68). In determining the moment when intervention would be required, it was concluded that:
It is not unlawful for the Secretary of State to decline to provide support unless and until it is clear that charitable support has not been provided and the individual is incapable of fending for himself such that his condition is verging on the degree of severity described in Pretty. (Q, EWCA, para. 119, viii)
This is the closest we come to a test in the CA judgment. According to Billings and Edwards (Reference Billings and Edwards2004: 98), the CA ‘finessed’ the application of article 3 restrictively, and the gap between NASS destitution and destitution for the purposes of article 3 was thus confirmed (dismissing a JCHR concern). In contrast to the JCWI case, where the legal tool available was to strike down the Regulations, attention was turned instead to degrees of suffering. We should also note that the emphasis lay not simply on withdrawal of support, but also on the prohibition of employment, which precludes any application of the ruling to the indigenous homeless.
The full impact of the decision on ‘treatment’ becomes clear in later cases, but in other respects the ruling delivered a politically nuanced judgment, which allowed both sides to claim victory.Footnote 7 Philips LJ criticised procedure and determined the meaning of treatment, but allowed the policy to function by maintaining the gap between NASS destitution and article 3. The judgment relied on the ECtHR for a definition of IDT, but without delivering a clear concrete test, setting the stage for further debate about the availability of charitable support and the acceptable degree of anticipation for provision under S55(5). Hence, in policy terms, a caution had been issued that enforced destitution where employment was prohibited could breach article 3, but a further eight judgments were required to resolve the matter of where the line should be drawn.
A fixed or bright line?
A step towards specifying concrete conditions was taken in Kay J's HC hearing on the case of S, D and T (EWHC 1941 (Admin) Hereafter S, D and T). Though one S55(1) decision was found to be flawed, the main burden of the judgment fell to addressing article 3 and applying the principles established in Q. Seeking a means to assess IDT, counsel representing the asylum seekers reformulated the Pretty definition into ten ways whereby treatment could breach article 3. This reformulation included a distinction between physical and psychological suffering, and was supported by a submission that dignity is the core value of our society. Lord Hoffman (Matthews v Ministry of Defense, para 26) was cited on the ECHR, to make the argument that, although social rights were not included in the convention, without the right to a minimum standard of living, many of the other rights would be meaningless.
It was also argued that the purpose of S55(5) was to prevent a breach of Convention rights, thus requiring anticipation.
The HO again relied on the ECtHR decision on O'Rourke, but Kay J dismissed the case as inapplicable since the claimant's suffering had been self-induced. Though the determining level of suffering was not explicitly addressed, Kay J did state that:
The Q test of ‘verging on’ connotes a degree of anticipation or, at least, a very fine temporal line. (S, D and T, para. 27)
On this basis, all three cases were found to qualify for support under S55(5), and in his closing paragraph, Kay J set out parameters for an IDT test:
Denied access to employment and other benefits, he will soon be reduced to a state of destitution. Without accommodation, food, or the means to obtain them, he will have little alternative but to beg or resort to crime . . . . In those circumstances and with uncertainty as to the duration of their predicament, the humiliation and diminution of their human dignity with the consequences referred to in Pretty will often follow within a short period of time. (S, D and T, para. 33)
The judgment thus moved closer to a concrete definition of IDT for S55(5) purposes. Humiliation and loss of dignity were critical, and they were placed in the context of contemporary values by reference to Simon Brown LJ's condemnation of ‘a life so destitute that . . . no civilised nation can tolerate it’ (JCWI, CA, p. 12).
This still left scope for interpretation in the application of the ruling, though of the three individual cases in S, D and T only T was appealed, yielding a judgment later criticised by the HoL. Although Q, EWCA had established the absolute nature of article 3, Kennedy LJ accepted an argument for restraint:
Because the allocation of resources is normally a matter exclusively for the executive, courts must be careful not to set too low the threshold at which the duty to act does arise. (T, para. 11)
This came close to turning an absolute right into a qualified right, to be determined by resource availability. Furthermore, HO counsel argued that a claimant must be able to present a stronger claim than the general mass of asylum seekers, or the situation would become unworkable, again a position more appropriate to a qualified right. The judgment was also significant in preserving the gap between destitution for NASS purposes and destitution which might trigger article 3 protection.
Eschewing any ‘simple test’ (T, para. 16), comparison was made between S and T. The case of S was recognised as inexorable – sleeping rough, forced to beg and suffering psychological disturbance and weight loss, he was verging on the required degree of severity. T had mental health problems he did not recognise, and was living (with some limited funds) at Heathrow in breach of local by-laws. The conclusion was that: ‘the boundary – which is not a fixed or bright line – lies somewhere between the two’ (T, para. 18), and T fell outside the scope of article 3. The case turned in part on a temporal issue and privileged the physical condition of the claimant above his loss of dignity. The first policy change followed some three months later, when the acceptable period for ‘as soon as reasonably practicable’ was extended from 24 hours to 72 (Refugee Council, 2004: 11).
Wait and see or common sense?
Despite the rulings thus far, the application of article 3 was still unresolved for want of a clear test of the level of severity required for a breach (cf. Cassese, Reference Cassese1991). The next four judgments demonstrate two contrasting approaches to how much suffering the application of section 55(1) could inflict before triggering article 3 protection. As we will see, the choice between them was critical in determining the feasibility of the policy in practice.
Zardasht
Newman J noted that a majority of asylum seekers share the facts of being homeless and without resources, of being strangers to the UK with no command of English and thus of being lonely, disorientated, anxious and vulnerable (R v SSHD on application of Zardasht, para. 5. Hereafter Zardasht). In M, it was precisely these features which qualified asylum seekers for ‘care and attention’ under the NAA. However, for article 3, Newman J applied a harsher test based on the extent of their effect; only in their ‘heightened presence’ would common features give rise to the ‘necessary degree of severity’ (Zardasht, para. 7). He added, ‘It may offend the social and moral values of others, but it is the law’ (Zardasht, para. 9).
There is no scope here for an expansionary view of rights, and, in contrast to other decisions, Newman J seemed to seal off the law from interpretation in terms of contemporary moral values. He reasoned that the purpose of S55 was to deny NASS support, which anyway required a test of destitution, and since individuals varied in their ability to cope, being destitute for weeks would not necessarily constitute a breach of article 3. The evidence in a case must show that the qualifying ‘state of severity’ is imminent (Zardasht, para. 19), with precise detail pertaining to food, shelter, health and hygiene. Furthermore, if there had been significant food deprivation:
there would be a measure of starvation and debilitation which any lawyer would have been alert to notice. (Zardasht, para. 31)
Newman J's ruling came to be termed the ‘wait and see’ approach; the intention of Parliament was to deny support, law took precedence over moral reservations, the features common to destitute asylum seekers were insufficient to make a case and the interpretation of fending for oneself was more restrictive than in the NAA. There must be discernible imminence of the Pretty criteria, and the ruling thus endorsed a policy that led to destitution, requiring state action only in extreme cases.
Limbuela
Collins J posed an alternative view in his HC judgment on Limbuela (R v SSHD on application of Limbuela. Hereafter Limbuela, EWHC). He reviewed evidence of the claimant's cough, pain, dizziness and feelings of trauma and humiliation, noting common sense suggests that sleeping rough was likely to exacerbate ill-health and could have further psychological effects. The claimant's attempts to access support were detailed, together with the denial of hostel accommodation for those with no recourse to public funds. This went against alleged availability of charitable provision, and led to the conclusion that:
If it is established that the claimant will be out on the street in winter, that no charity can assist him in that regard, and that although there is a possibility of some irregular food and washing, nonetheless, there is no real prospect of regular obtaining of such facilities, then the threshold is crossed. (Limbuela, EWHC, para. 30)
The judgment closed with a reprimand: a great deal of public money was being spent on court proceedings and lawyers fees, which a more flexible approach to S55 would reduce. In contrast with the Zardasht ruling, which applied only in extreme cases, the Limbuela case simply required evidence of no real prospect of support. The policy implications of the difference are profound, as the ‘common-sense approach’ all but closes the gap between NASS destitution and article 3 protection, stripping the policy of effect. Acknowledging his differences with Newman J, Collins J granted permission to appeal, but first came two further cases which weighed the relative merits of the wait and see and common-sense approaches.
Tesema
Tesema's case was critical in testing out the permitted degree of suffering, as it concerned a claimant who had not yet experienced time on the street. The issue was presented as a ‘test of the response of a civilised society to a pressing social problem’ (R v SSHD on application of Tesema, para. 19. Hereafter Tesema, EWHC) and echoing Kay LJ, Gibbs J implicitly placed the case in the terrain of social morals as well as law. The case was that the claimant was not fully fit, had experienced assaults and beating at home, was depressed and vulnerable and though not yet homeless was distressed at the prospect. In contrast to the O'Rourke case, Gibbs J stressed the involuntary nature of homelessness in S55 cases. He also noted the different ways in which the Pretty criteria could be met, implicitly drawing attention to mental suffering and to fear and anguish. He then weighed the ‘wait and see’ approach against the ‘common-sense’ effects of sleeping rough in winter.
Drawing on general principles beyond the letter of the law, he found it ‘contrary to any reasonable concept of justice’ (Tesema, EWHC, para. 59) to require someone to reapply for support when their condition had worsened, in the face of evidence that their basic needs could not be met. In contrast to Newman J, he also found social and moral values to be central:
The question of whether or not article 3 is infringed has to be determined on the basis of what a reasonable person, objectively applying the standards of a civilised society would find to be acceptable or otherwise. (Tesema, EWHC, para. 68)
The fact that Tesema had not yet slept rough was viewed as irrelevant to the fate he would face without support. The judgment also challenged the resource considerations implicit in Newman J's judgment:
Whatever the extent of the problem, it does not follow that a decision compelling an asylum seeker to live and sleep in the streets in winter is other than a breach of article 3. (Tesema, EWHC, para. 67)
The case was significant in finding the prospect of destitution rather than the actual experience to meet the required threshold.
Adam
In the subsequent case (R v SSHD on application of Adam, EWHC), Charles J began by noting that welfare issues are normally reviewed in the context of the statutory scheme, democratic accountability, efficient administration and the sovereignty of Parliament (Runa Begum v Tower Hamlets, para. 35). However, he also recognised (citing Q, EWCA, para. 115) that the more substantial the interference with human rights, the more the court would require by way of justification, and that the rights at issue were not confined to those of the ECHR, but also included the CSR and UDHR. A challenge to the S55(1) decision was quickly dismissed and the judgment accordingly turned to article 3. The claimant had been unsupported and sleeping rough for four weeks, though displaying only those features common to many asylum seekers. Nevertheless, it was judged that the duration of his predicament, the medical evidence and the absence of support were sufficient to meet the threshold.
Despite his emphasis on rights beyond the statute, Charles J nevertheless followed Newman J, finding Collins J's rejection of the ‘wait and see’ approach to display only fine distinctions from ‘real risk’, and therefore to be out of accord with earlier authority. Thus, two of these four cases endorsed the ‘wait and see’ approach and two the ‘common-sense’ approach. It remained for the CA and later the HoL to consider their relative merits, the point at issue being the degree of suffering which would require intervention from the Secretary of State, a matter which would determine the effectiveness of the policy itself.
Positive negative and absolute rights
The CA judgment on Adam, Limbuela and Tesema (R v SSHD on application of Adam, Limbuela and Tesema, EWCA. Hereafter Limbuela, EWCA) began with Laws LJ's dissenting judgment. This reopened the question of treatment in a manner that placed policy considerations to the fore. Laws LJ distinguished between article 3 breaches through violence by state servants, and violations stemming from acts or omissions by the state which expose the claimant to suffering; the key issue being the degree of direct state responsibility. There was a negative obligation on states to refrain from inflicting harm, but flexibility in article 3 to address other circumstances that might imply the need for protection and therefore a positive obligation. For Laws LJ only negative obligations were absolute, while positive obligations required a balance between ‘the general interests of the community and the interests of the individual’ (Limbuela, EWCA, para. 61). The effect was to undermine the absolute nature of article 3 in the light of policy objectives.
The legal reality was presented as a spectrum whereby in certain cases government action which caused suffering was justifiable. In deciding where the balance lay, account should be taken of the severity of the suffering, its origin and government purpose. Laws LJ recognised the ECHR as a living instrument, with an elastic quality, but argued that this flexibility should not be used to impose obligations beyond those originally envisaged in the Convention (Limbuela, EWCA, para 73)
Declining to act in the face of destitution, when there were no additional considerations, was not a breach of article 3 (para. 75). Having introduced government purpose into the equation, Laws LJ argued anything other than a ‘wait and see’ approach ‘would plainly emasculate the effect of S55(1)’ (para. 78). He thus gave policy factors considerable weight, and notwithstanding the 600 asylum seekers who would suddenly be without support (para. 79), allowed the HO appeal, with the admonition:
We cannot don the mantle of the statute's practical administrators . . . in order to save S55 from excoriation in the moral and political arena. (Limbuela, EWCA, para. 80)
This closing statement leaves hanging the role of moral principles in the interpretation of a statute or, as other judges have put it, an interpretation guided by the standards of civilised society though, in the event, Laws LJ held a minority view.
Carnwath LJ also adopted a policy focus, but to very different effect. He made reference to 650 cases awaiting a decision, the difficulty of applying the Pretty criteria and the absence of any single test that could resolve the issue. Q, EWCA was taken to establish that lacking charitable assistance and with no other means of fending for himself, a claimant would qualify for support without evidence of further suffering. However, Carnwath LJ queried other aspects of the decision in Q to argue that suffering by the destitute flowed from the absence of charitable support rather than treatment by the state. He therefore accepted the ‘spectrum’ analysis but nevertheless found a breach of article 3. In his view, the Convention required the state:
To take measures designed to ensure that individuals within their jurisdiction are not subjected to torture or inhuman and degrading treatment. (Limbuela, EWCA, para. 120)
For Carnwath LJ, this implied more than acting as a long stop in individual cases:
If the scale of the problem is such that individual breaches can only be avoided by more general action, such action can and must be taken ‘for the purpose’ of avoiding the individual breaches. (Limbuela, EWCA, para. 122)
At odds with his endorsement of the spectrum analysis, Carnwath LJ then asserted the absolute and fundamental nature of article 3, noting the very limited scope for deference to the Secretary of State (Limbuela, EWCA, para. 128). He found the approach of other judgments far too narrow for a problem that was systemic and related to the scale of numbers involved, such that individual solutions would not suffice. The third opinion, from Jacobs LJ, was on similar ground and also in agreement with the ‘spectrum analysis’, though finding this did not resolve the question of where on the spectrum a case fell. Jacobs LJ went further than Carnwath LJ in advocating a systemic approach:
One may not be able to say of any given individual that there is more than a very real risk that denial of food and shelter will take that individual across the threshold, one can say that collectively the current policy . . . will have that effect in the case of a substantial number of people. . .it must follow that the current policy is unlawful in violating article 3 (Limbuela, EWCA, para. 149)
Hence he offered an approach which, instead of focusing on degrees of suffering, a spectrum of accountability and individual rights, considered the collective impact of the regime.
The judgment overall found against the HO appeal and thereafter S55 was in effect suspended (Joint Committee on Human Rights, 2007, para. 90), except in maintenance-only cases. However, there was still no clear specification of a concrete test, though this eventually came as a result of a further HO appeal, to the HoL.
Resolution
In the HoL appeal (Adam, UKHL), the statements of caseFootnote 8 for each side show clearly contrasting orientations: for the HO, an interpretation of the ECHR in terms of its original focus, an assertion of the limits of positive obligations and an interpretation in the context of legitimate policy objectives and proportionality; for the claimants, an emphasis on the ECHR as a living instrument, on the absolute nature of article 3 and on a defence of fundamental rights over and above policy objectives.
Lord Bingham opened by recognising that the policy and purpose of S55 were a legislative choice and therefore not at issue. The question was rather what was required by the ECHR, and he accepted the ruling in Q which confirmed the regime as treatment for the purposes of article 3. On the IDT threshold, he recognised that, while many factors could affect a judgment, being compelled to sleep on the street, OR being seriously hungry, OR being unable to satisfy the needs of hygiene would qualify. A concrete test was thus finally provided, while Laws LJ's spectrum approach was dismissed by Lord Hope as having no foundation in the jurisprudence of the ECtHR. For Lord Hope, the application of S55(1) amounted to an intentionally inflicted act, which removed from asylum seekers the ability to fend for themselves, and required positive intervention as a remedy. Where the state was directly responsible for the action, this qualified as treatment, the prohibition was absolute and there was no question of balance with regard to legitimate government policy.
The word avoid was identified as the key to when preventive action was required; ‘it was not just a question of wait and see’ (Adam, UKHL, para. 62). Real risk was again deemed too generous, so ‘verging on’ and the ‘imminent prospect’ of a breach stood as the critical test. Lord Hope did not translate this directly into concrete indicators, but identified a safety net of wide reach, embracing all sorts of ‘inhumanity and degradation’, thus making implicit reference to dignity (Adam, UKHL, para. 59). Lord Scott agreed with the HO that treatment required more than failure, but he found that the whole regime imposed on asylum seekers could meet this test. Emphasis was again placed on the need to avoid a breach, and while in this, each case must be judged on its facts:
having to sleep out of doors would be a very strong indication that the threshold had been reached. (Adam, UKHL, para. 72)
Lord Scott also stressed not only physical suffering, but ‘growing despair and loss of self-respect’, and like Lord Hope made implicit reference to dignity. Similarly, Lady Hale read article 3 as a reflection of:
The fundamental values of a decent society, which respects the dignity of each individual. (Adam, UKHL, para. 76).
She also expressed unease with the spectrum approach, though recognising two types of state responsibility for suffering: through direct infliction of the suffering or through failure to provide protection. S55 was construed as a clear case of the first:
The state has taken the Poor Law policy of less eligibility to an extreme which the Poor Law itself did not contemplate, in denying not only all forms of state relief but all forms of self-sufficiency. (Adam, UKHL, para. 77)
Judged by the ‘standards of our own society’ (Adam, UKHL, para. 78), the result was deemed inhuman and degrading, and Lady Hale endorsed the Bingham criteria.
Finally, for Lord Brown state responsibility was the key, with no room for policy justification. While recognising that S55 necessarily contemplates homelessness for some, and that there is only limited scope for any positive obligation to house the homeless, Brown saw it as:
Quite another thing for a comparatively rich (not to say northerly) country like the UK to single out a particular group to be left utterly destitute on the streets as a matter of policy (Adam, UKHL, para. 99)
He emphasised the deliberate nature of the state's involvement, and quoted the Prime Minister on rough sleeping as a scandal in a ‘modern and civilised society’ (Adam, UKHL, para. 99), thus invoking contemporary moral standards. Doubting the practical impact of the policy, he found the consequences must be regarded as intended and therefore inhuman and degrading, or unintended and involving hardship disproportionate to the policy's aims. Either way, street homelessness crossed the article 3 threshold.
The decision was unanimous, and the issue on which the judgment turned was the significance of state responsibility in engaging article 3. A rejection of the spectrum approach and the involuntary nature of the situation confronting late claimers confirmed S55 as treatment, and human dignity was endorsed as a central principle. The use of article 3 in these rulings thus strikes a contrast to the tradition of judicial deference in immigration law, and serves as a positive illustration of the impact of the HRA (cf. Stevens, Reference Stevens2004). While there were references to the non-political nature of the judicial role (Adam, UKHL, paras 2, 14, 85), these were partially belied by indirect statements of moral censure (Adam, UKHL, paras 66, 77, 99), and an awareness of the broader significance of the polarised positions represented by the case (Adam, UKHL, paras 13, 53, 73, 101).
The politics of judgment
Taken together, the cases examined here provide a forceful illustration of the power of rights in challenging policy interests, and demonstrate the significance of general principles or background values in their interpretation (Dworkin, Reference Dworkin1977). However, not all judges draw equally on such principles, and the present history reveals a distinctive pattern; the more expansive interpretations occur where judges take recourse to general principles, while restrictive interpretations are more closely confined by the words of a statute and policy imperatives. One role of the cases documented has therefore been in determining an appropriate framework, with judges variously making reference not only to justice, fairness and dignity, but also to the standards of civilised society, decency and common sense. In so doing, they place their judgments on a policy of deterrence by destitution in the context of broader social values, which have a powerful bearing on the boundaries of social and moral obligation.
Dialogue has played a significant role in establishing this framework (cf Habermas, Reference Habermas1996), both in terms of judicial exchanges within and between cases, and in political dialogue beyond the courtroom, manifest in the strategic delivery of many judgments. This delivery may be ameliorative, with judgments protecting the coherence of legislation, or offering assurance that policy is viable if procedures are amended. It may emphasise government resource constraints or reprimand government profligacy and so on. But delivery can also underline issues of principle, as in critical comment on the lack of parliamentary scrutiny, the rationale of policy, its moral legitimacy and even grounds for future challenge. In recognising these possibilities, we begin to see the functioning of the judiciary not just in terms of their structural position in a system, but also as actors in a social and political exchange. The whole sequence of cases can be viewed not simply as the pursuit of legal reason, but also as an inherently political conflict (Unger, Reference Unger1983) about the application and interpretation of the general principles which must govern policy choices.
Before closing, however, a note of caution is due. While the judgments documented here led to the effective suspension of section 55, except in maintenance-only cases or where access to resources is clear, the legislation remains intact. Furthermore, the rulings have no purchase on the indigenous homeless, and would read differently were asylum seekers granted a right to employment. Nor do the judgments affect the many failed asylum seekers living in destitution, but unwilling to return to their country of origin (Guardian, Comment and Features, 2008: 4); destitution remains a legitimate policy tool in the government's dealings with this group, and may prove much harder to overturn.
Acknowledgement
I thank the British Academy/Leverhulme Trust for supporting this work, and Robert Fine for generous comment and discussion.
