R (Lumba) v Secretary of State for the Home Department concerned claims by foreign nationals who were due to be deported upon completion of their prison sentences for various crimes. The men were detained on the authority of the Home Secretary under paragraph 2(2) of Schedule 3 to the Immigration Act 1971 prior to their deportation. Mr Lumba, a Congolese national, was convicted of various offences including wounding with intent, and was sentenced to a term of four years’ imprisonment. The day before his scheduled release, he was notified that the Secretary of State had authorised his detention pending deportation. Mr Mighty, a Jamaican national, was imprisoned in respect of offences including possession with intent to supply a Class A drug. The Secretary of State had a published policy on the detention of prisoners pending deportation, which referred to a ‘presumption’ in favour of release. However, it transpired that a very different policy had been applied in the relevant period, which was a ‘near blanket ban on release’ (Lumba 2011, para 5, Lord Dyson JSC, quoting a document from the then Home Secretary). The claimants were detained pursuant to this secret, unlawful policy, and brought claims in false imprisonment . However, it was accepted that, had the Home Secretary applied the published policy, the men would still have been detained: ‘each claimant had a very bad criminal record and would undoubtedly have been kept in custody under the Secretary of State’s published policies.’ (Lumba 2011, para 195, Lord Walker).

There were two principal points of relevance¹⁴ for the Court to address. First, given that the inevitability of the men’s detention had the published policy been applied to them was accepted, did that mean that their detentions were lawful? This is the ‘causation test’ question. Second, if they were able to recover in principle, what was the extent (or nature) of the damages to which they were entitled? This is the ‘damages’ question.¹⁵ The Supreme Court divided 6:3 on each point, but with different majorities.

Two months after the decision in Lumba, the Supreme Court considered a variation on these themes in R (Kambadzi) v Secretary of State for the Home Department. The case had been heard by a five-Justice panel before Lumba, but judgment was delayed in Kambadzi so that the parties could make submissions on the basis on the reasoning in Lumba. The appellant, Mr Kambadzi, had been convicted of common and sexual assault. He was sentenced to a year’s imprisonment, and before he became eligible for release a deportation order was made against him. Thereafter, his detention had not been subject to sufficiently regular review as required by rule 9(1) of the Detention Centre Rules 2001 and the Secretary of State’s published policy.¹⁷ Without good reason, the necessary reviews were either not carried out at all or not carried out correctly. The appellant sued in false imprisonment , arguing that his detention had been unlawful because of the failure to comply with this procedure.

The Court was divided, along the same lines as in Lumba. The majority, Lord Hope DPSC, Baroness Hale and Lord Kerr, had been in the majority on the causation test point in Lumba. We may note that, although Lord Hope and Baroness Hale were in the minority on the damages point in Lumba, it was accepted that, in the light of that decision, there was no prospect of establishing an entitlement to vindicatory damages in Kambadzi. Lords Brown and Rodger again dissented.

For the majority, the review was a prerequisite of the continued authority to detain. ‘Unless the authority to detain was renewed under the powers conferred by the statute [the appellant] was entitled to his liberty’ (Kambadzi 2011, para 54, Lord Hope). Without it, the detention was unlawful. On the other hand, ‘the continued detention could at all times have been justified by the Secretary of State had he been faced with an application for judicial review’ (Kambadzi 2011, para 34, Lord Hope). The position is succinctly summarised by extracting from two paragraphs of Lord Kerr’s judgment: ‘causation is not a necessary ingredient for liability’ but ‘is relevant to the question of the recoverability of damages’ (Kambadzi 2011, paras 88–89). The possible conclusion that only nominal damages should be awarded ‘does not affect the issue of principle’ (Kambadzi 2011, para 56). Furthermore, Lord Hope declined to ‘foreclose… the possibility that the appellant [may be] entitled to more than a purely nominal award’ (Kambadzi 2011, para 55).

Again, for the minority, Lord Brown was almost incredulous in his dissent. His judgment contains a reassertion of his reasons in Lumba and his belief that the outcome ‘devalue[s] the whole concept of false imprisonment’ (Kambadzi 2011, para 108). (His Lordship noted that the first 23 paragraphs of his judgment were an expression of his view before the decision in Lumba). But Lord Brown was in any event of the view that the majority approach in Lumba did not dictate that the claimant must succeed in Kambadzi. Mr Kambadzi had been lawfully detained in the first place; the only defect here was a procedural one in relation to the continuing review. As in Lumba, the dissent turns on a different view of the status of the counter-factual that, had the reviews been correctly carried out, the claimant would still have been detained. The majority, as we have seen, held that the lawfulness of the claimant’s detention needed to be re-established at each review. The dissentients by contrast view the procedural requirement as giving the detainee the opportunity to establish that they should be released. For that reason, Lord Brown viewed it as ‘still more undesirable’ to find liability in Kambadzi than it had been in Lumba (Kambadzi 2011, para 115). Indeed, the implications of the respective approaches in Lumba for a case such as Kambadzi were expressly considered as hypothetical examples given by some of the Justices in the former case.¹⁸ And, in the case of the minority, there was the suggestion that the case would be less compelling: Lord Phillips viewed the ability to recover in a case of a failure to review the detention to be an ‘extraordinary result’ (Lumba 2011, para 336).

These two cases demonstrate confusion over how, if at all, ‘causation’ is relevant to liability in false imprisonment . Our focus tends to be on loss when discussing causation: ‘[tort] liability in general is standardly thought to require successful application of the “but for” test of causation in fact’.¹⁹ But on the majority view of false imprisonment, causation is only applicable insofar as it is established that the defendant ‘caused’ the imprisonment in a narrow sense: to quote from Bird v Jones (1845, 115 ER 668), ‘the acts will amount to imprisonment neither more nor less from their being wrongful or capable of justification .’²⁰ The counter-factual problem consequently raises questions about the essence of the tort of false imprisonment and a rights-based analysis of the wrong, as it is not clear that we have a counter-factual baseline at all.²¹

If the ‘causation test’ is not to operate as a control mechanism, we have to reflect on how we attribute responsibility and impose liability in the tort . It must be recognised that there remains some uncertainty as to the degree or nature of public law error that will ‘justify resort to the common law remedy’ (Kambadzi 2011, para 40; Steel 2011, p. 529). Indeed, one of the public errors in Kambadzi was deemed insufficient to establish compelled the conclusion that there be liability in such cases for one month in December 2007 the review was ‘carried out by an official of the wrong grade’ (Kambadzi 2011, para 60).

The analysis above has shown that the counter-factual or fictional scenario is central to the cases. Underlying the debate in the Lumba and Kambadzi, especially in the minority judgments and the majority view on the damages point, are apparent misgivings about seeming to ignore the ‘fact’ (or ‘counter-fact’) that the claimant would still have been imprisoned. It is a more substantial concern than the mere feeling that the given appellant ‘may not be a very nice person’ (Kambadzi 2011 para 61, Baroness Hale). And most of the Justices clearly believed that the counter-factual ought to be taken into account in some way.²² Even the minority on the damages points accept that the claimants could not show that they had suffered any loss: hence the perceived need to identify the special character of the wrong with a novel award of vindicatory damages.²³ The Justices’ divergent analyses on the two points are then ultimately based on an acceptance that, at least in some sense, the alternative scenario is relevant to the resolution of the case.

The case law from Fairchild onwards has concerned claims by victims of mesothelioma, which is an invariably fatal cancer caused principally by exposure to asbestos. The claimants in Fairchild itself had been exposed to asbestos while working for more than one employer and had developed mesothelioma. Scientific knowledge did not enable the determination of the source of the causative asbestos, because the aetiology of mesothelioma is uncertain (and the precise understanding of the uncertainty has itself evolved in parallel with the litigation).²⁹ Thus, where a claimant had been exposed to asbestos in breach of duty by more than one employer (only one of whom may now be solvent and worth suing), and subsequently developed mesothelioma, it was impossible to identify the source of the particular exposure which had caused the cancer. There was an ‘evidentiary gap’ and the claimant could not prove causation according to the ‘but for’ test, because it was possible that the causative asbestos fibres had come from another employer.

Their Lordships unanimously held that in such a case the claimant should be entitled to succeed, on the basis that the defendant had ‘materially increased the risk’ of the harm eventuating, relying upon McGhee v National Coal Board