public law,

Privacy International: a constitutional barometer

By Vasiliki Poula Aug 20, 2019

In the 1960s, there was talk of a ‘Fairness Revolution’ in English judicial review.1 One of the most pertinent cases exemplifying such a movement is Anisminic.2 In that case, an administrative statutory body, the Foreign Compensation Commission, was given the job of deciding whether compensation should be awarded for property sequestrated as a result of the 1956 Suez crisis. This was empowered by section 4(4) of  the Foreign Compensation Act 1950,3 which provided that:

‘The determination by the Commission of any application made to them under this Act shall not be called in question in any court of law.’4

On the face of the provision, there seems to be very little, if any, room for judicial maneuver: if no ‘determination’ of the Foreign Compensation Commission could be ‘called in question’ by any court, then it is difficult to see what scope might remain for judicial review.

Yet, judges found their way around this seemingly exhaustive provision. Since any determination of the Commission should not be scrutinised by courts, judges would be able to do so only if it was found not to be a determination. So, the only way that they could enforce their constitutional role would be to prove that what the Commission called a ‘determination’ was not a real determination.

The question consequently turned to whether the Commission actually made a determination. To answer the question, they recalled an error of law that the Commission had committed made in misconstruing the eligibility criteria that had to be met for the claimant’s compensation. More specifically, when determining that the appellant did not qualify for compensation, the majority considered that the Commission had gone outside its jurisdiction by basing its decision on an inquiry into the nature of the appellant company which the legislation did not empower it to make.5

As such, what the Commission regarded as a ‘determination’ did not qualify as a determination for the courts – instead, they called it a ‘purported determination’, a ‘nullity’, because the Commission could not be held entitled to make procedurally unfair determinations.6 The House of Lords concluded that the provision was not enough to oust judicial review for error of law.

Fast forward 50 years, and another similar case emerges, Privacy International.7 The issue was whether the Investigatory Powers Tribunal (IPT), a body set up to review state surveillance decisions, was itself subject to judicial review in the ordinary courts. The case concerned the legal scope of s.68(7) of the Regulation of Investigatory Powers Act 2000,8 which provided that:

‘determinations, awards, orders and other decisions of the Tribunal (including decisions as to whether they have jurisdiction) shall not be subject to appeal or be liable to be questioned in any court’.9

In particular, the question was whether this provision constituted a successful attempt to oust the jurisdiction of the High Court to hear challenges to the decisions of the Investigatory Powers Tribunal by judicial review.

In light of the recent Supreme Court decision, the case has come to be treated as an Anisminic 2.0.10 Yet, it is significant to remember that the Court of Appeal decision had a different view, having marked the two cases as materially different.

Sales LJ found two differences. Firstly, there was a difference in the language used in the two provisions. This was because the drafter of section 67(8) had expressly referred to the possibility of the IPT making an error of law going to its jurisdiction or power to act, by the words in parentheses in that provision: ‘including decisions as to whether they have jurisdiction’.11 Therefore, at least so far as the word ‘decision’ was concerned, it was not tenable to apply the simple distinction relied upon in Anisminic in the context of section 4(4) of the 1950 Act between a determination and a purported determination, in the sense of a determination made without jurisdiction because of an error of law. Therefore, to use the jargon of Anisminic, section 67(8) applied to purported decisions, too.

Secondly, there was a difference in the context of the two cases, since the provisions applied to different bodies. By contrast with Anisminic, the tribunal of Privacy International was of a judicial nature, already exercising a supervisory jurisdiction and power of judicial review over the actions of public authorities.12 Furthermore, the legislation authorised the Secretary of State to create a right of appeal (although that power had never been exercised).13 Sales LJ also noted the very high quality of the IPT in terms of judicial expertise and independence,14 since to him, it was clear that Parliament’s intention in establishing the IPT and in laying down a framework for the special procedural rules which it should follow, was to set up a tribunal capable of considering claims against the intelligence services under closed conditions. The ultimate goal was to have complete assurance that there would not be disclosure of sensitive confidential information about their activities.15

Nevertheless, there are counterpoints to both of those issues. On the first linguistic difference, an adequate response is given by Leggatt LJ, who while not formally dissenting, was ‘inclined’ to a different view. He thought that the case was governed by the reasoning in Anisminic, since the only potentially relevant difference in the wording of section 67(8) is that it contains the words in parentheses ‘(including as to whether they have jurisdiction)’. But he found it hard to see how these words could make a critical difference in the light of Anisminic. This is because the decision in Anisminic was not concerned by its scope of jurisdiction, i.e. the tribunal was not found to make an error of law, because it made an error about the scope of its jurisdiction.16 Rather, it decided that any determination based on an error of law, whether going to the jurisdiction of the tribunal or not, was not a ‘determination’ within the meaning of the statutory provision. Therefore, the question of jurisdiction emerges as irrelevant.

On the second contextual difference, we should first examine the policy justifications behind it. Sales LJ found the status of the tribunal significant, because he noted that if judicial review was to be allowed, then – and he went on to use a floodgates argument – it would be open to a claimant to allege that a decision had been made, unsupported or contradicted by evidence, requiring the reviewing court to examine all the evidence which was before the decision-making body.17 But as David Hart QC notes, prior to the Supreme Court decision, ‘the underlying issue in the case (which is ripe for consideration in the Supreme Court) is whether the practical protections built into RIPA for the benefit of the Security Services should be regarded as overriding the skepticism which the courts have expressed about any attempt to oust their powers.’18 And indeed, the Supreme Court found such practical considerations less important in relation to the rule of law preoccupations that emerged. After all, as Leggatt LJ pointed out, it is striking that no language so far used has been held to be sufficiently clear to oust judicial review,19 and Privacy International did not become a first.

Profound constitutional concerns are embedded in both of those arguments that purport to parallel Anisminic with Privacy International, as the mirroring of the ruling of the first on the latter would lead to the re-evaluation of fundamental notions; when Privacy International reached the Supreme Court, everyone knew that the resulting judgments would constitute a useful barometer of where thinking lies at the highest judicial level on the most delicate of constitutional issues, more than ten years on from Jackson20 and two years after Miller21. It is neither easy nor particularly productive to attempt a clear demarcation between the various constitutional concerns, but I would like to focus on statutory interpretation, parliamentary sovereignty, the rule of law and the idea of a constitution.

There have been numerous attempts to give a definitive answer to what statutory interpretation entails. Yet, Lady Hale in R (Black) v Secretary of State for Justice22 seems to succeed in verbalizing a well-established rule, namely that ‘the goal of all statutory interpretation is to discover the intention of the legislation’, and ‘that intention is to be gathered from the words used by Parliament, considered in the light of their context and their purpose’.23

Returning to the case of Privacy International, what does the intent of the legislature seem to be? The words used by Parliament in the ouster clause of Privacy International are different from those of Anisminic. This difference existed for a reason, and it seems logical to infer that Parliament wanted to avoid a similar ruling with that of Anisminic, as it is clearly a more elaborate attempt to exclude judicial review. This observation is not mutually exclusive with Leggatt LJ’s attempt to establish that the language of the ouster clause of Anisminic might be different, but not materially different from that of Privacy International. But from those two views, do we conclude that Parliament did not intend ‘strongly’ enough to oust judicial review, and that is why the wording was different, but not materially different? This interpretation of the statute would lead to the maintenance of the judicial review. Or do we infer that Parliament did intend very strongly to oust judicial review, but the use of ‘jurisdiction’ did not have the same legal effect as if the term  ‘error of law’ might have been used, which would have been an even more express declaration leaving no space for maneuver. This interpretation of the statute would lead to the ousting of judicial review.

The court decided that the first path was more appropriate. There was no direct address of this dilemma, but their preference became evident when they noted that the exercise is not one of ordinary statutory interpretation, as there is a common law presumption against ousting the jurisdiction of the High Court: the plain words of the provision must yield to the principle that it will not protect a decision that is legally invalid.24 It is welcome that the Supreme Court has reaffirmed the strictness of the presumption against ouster, in light of the Court of Appeal ‘seeing force’ in the submission that the principle of legality is of ‘variable impact’. This presumption, though, is in effect a disguised prohibition set and enforced by the courts. Therefore, the well-established legislature-centric schema of statutory interpretation emerges as pro tanto: the court will give effect to the legislative intent as long as it is in line with the judiciary’s intent.

In confirming an alternative approach to statutory construction, a challenge to parliamentary sovereignty is born, as the Supreme Court appears to be challenging the legislature’s legally unlimited law-making authority. The Supreme Court divided their decision into two parts, and the constitutional doctrine of parliamentary sovereignty was implicated in each strand in fundamental ways: first, in the specific context of interpreting Parliament’s legislative intention in enacting a provision which purported to immunise decisions of the Investigatory Powers Tribunal from further legal challenge; and second, in relation to whether it can ever be legally possible, in general, for Parliament to enact an ouster clause of this kind.

On the one hand, such speculation about this doctrine of parliamentary sovereignty could be treated as non-determinative of the scope of parliamentary legislative authority, for if Parliament truly is sovereign, this is a matter the courts do not have the authority to decide.25 Instead, the case could be perceived as revealing in so far as it provides a clear insight into the constitutional framework within which many judges increasingly perceive their role and their ability to express ideas of legislative authority.

On the other hand, what if courts are indeed trying to decide the future of parliamentary sovereignty? Recent case development would substantiate this hypothesis. As Lord Steyn notes in Jackson, ‘the judges created this principle [of parliamentary sovereignty]. If that is so, it is not unthinkable that circumstances could arise where the courts may have to qualify a principle established on a different hypothesis of constitutionalism.’26 In Privacy International, we see judicial acceptance of the notion that there might be specific limits on Parliament’s power, which come not from EU law, as was the case in Factortame (No 2), Thoburn or Miller but from the domestic constitutional principle of the rule of law.

Moving on to the rule of law concern, it should be noted that this concept was found in the nucleus of the second question to which the court turned: were there circumstances under which the courts would decline to give effect to a statutory provision which, even when interpreted in light of the ‘fundamental presumption’, did purport to wholly oust the supervisory jurisdiction of the ordinary courts? In Privacy International, it was claimed that the constitutional principle of the rule of law had certain fundamental requirements which no ouster clause, no matter how clearly expressed, could deviate from.27 Although their Lordships did not decide on the general lawfulness of ouster clauses, Lord Carnwath sees a strong case for holding that, ‘consistently with the rule of law, binding effect cannot be given to a clause which purports wholly to exclude the supervisory jurisdiction of the High Court to review a decision of an inferior court or tribunal, whether for excess or abuse of jurisdiction, or error of law’.28

As Leggatt J put it, there is extreme reluctance to ‘attribute to Parliament an intention to achieve a result which would be so clearly inconsistent with the rule of law’,29 using language which reminds us of Lord Hope’s words in Jackson, where he claimed that ‘parliamentary sovereignty is no longer, if it ever was, absolute […] The rule of law enforced by the courts is the ultimate controlling factor on which our constitution is based.’30 The full significance of those statements, which was not part of the binding ratio, may be discovered if the day comes that Parliament passes an ouster clause extreme enough to embrace such an eventuality. Of course we should not forget that there should be a choice: we are not compelled to accept the structures being developed by the courts, which are by no means immutable or inevitable.  Or even if we accept them, it is important for them to have democratic legitimisation by political mechanisms – Acts of Parliament might be made by here-today, gone-tomorrow politicians, instead of the Burkean latent wisdom of the common law, but it is these same politicians that hold the democratic mandate.

But the indisputable effect of having such legal principles developed in isolation from the political arena, over and above political (and especially democratic) principles, is that in some contexts, courts appear to gain ultimate constitutional authority, while in other contexts, parliamentary sovereignty still emerges as uncompromised. Professor Gordon attempted to capture the nature of this ‘catch-all’ constitutionalism by using the term ‘synthetic constitution’, i.e. a framework which is superficially plausible, perhaps in some ways even attractive, but equally one which we are not compelled to accept as absolute or authentic.31 Independently of how ‘artificial’ such an approach is, this ‘synthetic constitution’ seems to be the modern constitutional settlement. As such, maybe we are moving towards a new hypothesis of constitutionalism, which transcends the labels of ‘political’ or ‘legal’ as mutually exclusive and instead, attempts to combine the two in an optimal way.

[1] An interesting read comparing the Fairness Revolution of the 60s with a modern reconfiguration of the law of judicial review is: Thomas Poole, ‘The Reformation of English Administrative Law’ [2009] CLJ 68(1) 142.

[2] Anisminic Ltd v Foreign Compensation Commission [1969] 2 AC 147.

[3] Foreign Compensation Act 1950.

[4] ibid s 4(4).

[5] Anisminic Ltd v Foreign Compensation Commission [1969] 2 AC 147 [195] (Lord Pearce).

[6] ibid [153].

[7] R (on the application of Privacy International) v Investigatory Powers Tribunal and others [2019] UKSC 22.

[8] Regulation of Investigatory Powers Act 2000.

[9] ibid s 68(7).

[10] J Metzer, ‘Anisminic 2.0’ (UK Human Rights Blog, 15 May 2019) <> accessed 5 August 2019.

[11] R (on the application of Privacy International) v Investigatory Powers Tribunal and others [2017] EWCA Civ 1868 [34] (Sales LJ).

[12] R (on the application of Privacy International) v Investigatory Powers Tribunal and others [2017] EWHC 114 (Admin) [42].

[13] ibid [43, 45].

[14] ibid [38].

[15] ibid [42].

[16] R (on the application of Privacy International) v Investigatory Powers Tribunal and others [2017] EWCA Civ 1868 [55] (Leggatt LJ).

[17] ibid [44].

[18] David Hart QC, ‘Can you draw a line between this case and Anisminic?’ (UK Human Rights Blog), 25 November 2017 accessed 5 August 2019.

[19] R (on the application of Privacy International) v Investigatory Powers Tribunal and others [2017] EWHC 114 (Admin) [52].

[20] R (Jackson) v Attorney General [2005] UKHL 56.

[21] R (Miller) v Secretary of State for Exiting the European Union [2017] UKSC 5.

[22] R (on the application of Black) v Secretary of State for Justice [2017] UKSC 81.

[23] ibid [36] (Lady Hale).

[24] R (on the application of Privacy International) v Investigatory Powers Tribunal and others [2019] UKSC 22 [15].

[25] Mike Gordon, ‘Privacy International, Parliamentary Sovereignty and the Synthetic Constitution’ (UK Constitutional Law Association, 26 June 2019) <