the Afghan leak super-injunction case as a bonfire of constitutional principles – UK Constitutional Law Association

Introduction

On Tuesday last week it emerged that a mind-boggling data breach by the UK government – and the momentous policy decisions taken in its wake – had been kept a secret from all but a handful of people across the three institutions of the state for nearly two years.

In a nutshell, in August 2023 the UK government became aware of an enormous Ministry of Defence (MoD) leak involving the personal details of over 18000 Afghan nationals who assisted the UK during its military operations in Afghanistan, as well as details of some of their UK sponsors. Whether the Taliban was aware of the leak or not was unknown: if it became aware, and those details came into its possession, the lives of many on that list might have been put in grave danger.

Therefore, and on the back of initial media interest in the story, the then government sought an injunction contra mundum (against all) to keep the leak a secret, ostensibly to guarantee the safety of those on the list while emergency remedial action was taken. But a single King’s Bench Division judge, staffing the court during the summer holiday period, went beyond the government’s request and granted  an unprecedented* (vis-a-vis the government) super-injunction: a court order which prohibits disclosure not only of the underlying information, but also of the existence of the court order itself. Fast-forward two years, and five judicial decisions by two different senior courts, later.

Extremely helpful summaries of the timeline of events – including the series of public and ‘closed’ judicial hearings and decisions that followed – can be found here and here. There is a huge and diverse range of public law issues to be unpacked, including the disputed choice of procedure before the High Court; the growing scepticism about the level of scrutiny of the government’s evidence by courts in ‘closed’ proceedings involving national security; and the impact that super-injunctions have on freedom of expression and media freedom. In this post, I will focus on what I take to be some of the main (but certainly not all) constitutional implications of these proceedings.

On the ‘super’ element of the injunction

The first thing to note is that Robin Knowles J, who first heard the MoD request for an injunction contra mundum, granted an even more problematic super-injunction of his own volition, without much in the way of justification for doing so.

As others have already remarked, no one can deny Knowles J was confronted with an extremely difficult choice when presented with the government’s request for an injunction, given that the lives of so many were potentially at stake. Arguably, however, this context would have called for an explicit and full justification for going beyond the MoD request, rather than implying that the ‘super’ element merely followed from the justification for the underlying injunction [at 10-11]. This is so for two reasons in particular.

First, once the question becomes whether the ‘super’ element of the injunction should be lifted, rather than whether it should be granted in the first place, the burden of justification shifts from the government to the other parties in the proceedings. Second, once the ‘super’ element of the injunction had been established, discharging it would have created a ‘real risk’ that the protection afforded to the underlying information could be ‘unravelled’, as Chamberlain J notes in his first judgment [at 44-46] and as was confirmed by the Court of Appeal [at 72-78].

Therefore – despite that for Chamberlain J the strength of this consideration had been defeated by other factors by the time of his third judgment in May 2024 – it appears that the ‘super’ element of the injunction became its own justification.

The lack of explicit and robust justification by  Knowles J for going beyond the MoD request is disappointing. Not only did it make his reasoning for granting an unprecedented*, contra mundum, super-injunction to the government hard to discern, but it also arguably compounded the difficulty faced by Chamberlain J in evaluating whether to keep the ‘super’ element of the injunction in place or not once he took over the case. Indeed, one cannot  but wonder whether it would have been preferable if, notwithstanding the summer holiday period, the initial injunction request had been heard by Chamberlain J himself, or by another specialist judge from the Media and Communication List (these are High Court judges with specific experience in all types of media cases).

This suggestion, it should be noted, runs against the finding of the 2011 Report ‘Super-Injunctions, Anonymised Injunctions and Open Justice’ by the Committee on Super-Injunctions, led by the then Master of the Rolls, Lord Neuberger. In that report, the Committee warned against the exclusive ‘use of specialist judges to hear applications for interim non-disclosure orders, including super-injunctions and anonymised injunctions’. But a comparison between the care and sensitivity displayed by Chamberlain J throughout this case to its constitutional implications, as opposed to that showed by other judges that were involved, appears to provide a compelling counterexample to the conclusions on this point in the Report on super-injunctions.

Parliamentary accountability as an afterthought: the CoA overturns Chamberlain J’s decision to lift the super-injunction after 8 months

Perhaps the most conspicuous omission from the initial decision to grant the super-injunction, and especially from the Court of Appeal (CoA) decision in July 2024 to overturn the decision by Chamberlain J to finally lift the injunction in May 2024, is that of any reference to the constitutional principle of parliamentary accountability, as famously identified in 2019 by the Supreme Court in Miller II. This glaring omission was promptly noted after the publication of the judgments on the 15^th of July, including by Mark Elliott in his characteristically excellent quickfire analysis of the case.

Chamberlain J, unlike the other judges involved in this case, seems acutely aware of the effects of the super-injunction on parliamentary accountability. It is worth reproducing in full paragraph 16 of his first judgment in November 2023 (emphasis added):

‘I should explain that the injunction did not constrain what could be said in Parliament. Under Article IX of the Bill of Rights, no such constraint would be constitutional or lawful. I varied the injunction to make this point clear for the avoidance of doubt: see paragraph 15 of my order of 13 October 2023. Nonetheless, MPs and peers cannot ask questions about something they do not know about; and the Parliamentary authorities may regard the existence of an injunction a relevant [sic] to their decisions about what can and cannot be raised.’

It is astonishing, then, to see the CoA giving no special weight to the principle of parliamentary accountability when it overturned – 11 months after the initial superinjunction was granted and after the general election had taken place in the meantime – the decision to lift the super-injunction by Chamberlain J. As Elliott has charitably put it, this ‘reflects a highly impoverished view of the principle of parliamentary accountability’. This criticism of the CoA’s decision might be reinforced by at least two further observations.

The super-injunction and parliamentary accountability: One, No One and One Hundred Thousand?

First, the model of parliamentary accountability implied by the government request for an injunction, the decision to grant the super-injunction by Knowles J, and the decision by the CoA, appears to be a ‘monolithic’ one: either all members of the Houses of Parliament are made aware of the leak, or none of them is. And given what was (and to some extent still is) at stake in this case, it seemed reasonable to conclude that the right choice was indeed the latter (at least in the short term).

This is, however, a false dilemma, given that parliamentary accountability comes in many forms. And while it is, conceivably, only the sum of them all which constitutes the basis of the Westminster model of democratic government, I cannot think of any reason why the government should have not sought to facilitate selected forms of parliamentary accountability while it was assessing how best to proceed in light of the leak. The Intelligence and Security Committee of Parliament (ISC) is the first and foremost example: its remit is national security, its members have security vetting, are subject to Section 1(1)(b) of the Official Secrets Act 1989 and are routinely given access to highly classified material in carrying out their duties.

Unsurprisingly, members of the ISC are reportedly ‘furious’ that the information was withheld from them. But other ways could have been sought to allow parliamentary oversight while maintaining the necessary secrecy around the leak: it was Chamberlain J himself, in this first decision to renew the super-injunction for an additional four weeks in November 2023, who suggested [at 46] the government might seek parliamentary oversight either through the ISC  or via engagement with the opposition ‘on Privy Council terms’ (that is, with the understanding that the confidential information shared will remain secret).

The fact that none of this happened is, once again, extraordinary, given the scale of the public policy and financial decisions that needed to be made in light of the leak, and compounds the impression that the CoA wholly failed to give due weight to Chamberlain J’s careful reasoning in lifting the super-injunction after 8 months, in May 2024.

Shhh… it’s a secret: how the super-injunction puts the Speaker in a constitutional conundrum

It is understood that, at some point around December 2023, the Speaker of the House of Commons was made aware of the leak and the resulting super-injunction. Save for in respect of parliamentary proceedings, which are protected by Article IX of the Bill of Rights, he would presumably therefore also have been bound by the super-injunction, in much the same way as journalists and other politicians who became aware of these facts.

As Baroness Harman has suggested, the Speaker would likely have been informed about this super-injunction (as with others) against the backdrop of Parliament’s sub judice rule. This internal rule (formalised most recently in a 2001 Commons resolution) is intended to avoid, or at least to minimise, the risk of parliamentary proceedings prejudicing the outcome of live court cases. While the sub judice rule does not restrict debate on legislation nor ‘when a ministerial decision is in question’, and can be relaxed by the Speaker in relation to specific parliamentary proceedings, it is only exceptionally departed from, and inhibits parliamentary scrutiny about the specifics of a case.

The Speaker of the House obviously could not enforce the sub judice rule in relation to this case unless he was himself aware of the court proceedings. Not informing him would have risked him (or another occupant of the Chair) being blindsided if and when an MP raised the matter on the floor of the House, or in other parliamentary proceedings. From this perspective, informing him helped to preserve the “comity” between the judiciary and the legislature.

But disclosure to the Speaker was not without consequence. As Chamberlain J himself acknowledged in his first judgment [at 16, emphasis added]:

“MPs and peers cannot ask questions about something they do not know about; and the Parliamentary authorities may regard the existence of an injunction a relevant [sic] to their decisions about what can and cannot be raised.”

Even though the super-injunction did not impose legal limits on parliamentary accountability, it was clearly intended to impose practical, procedural and political ones. The court seems to have been relying on the Speaker to enforce the sub judice rule so as to preserve (so far as possible) secrecy in relation to the leak and the super-injunction. This posed, for the Speaker of the House, a clear conflict between two constitutional principles: (a) comity with the courts and (b) parliamentary accountability. The Speaker’s primary constitutional role, after all, is to defend the interests and privileges of the House of Commons, so that it can effectively debate matters of public interest and hold to account the executive.

Such a tension might be tolerable in the very short-term where national security considerations, and the lives of affected Afghan nationals, were believed to be at stake. But the longer the super-injunction applied, the less tolerable this constitutional conflict would be. The opportunities for parliamentary accountability (a) for the leak, (b) for the government’s policy response and (c) for the appropriateness of the super-injunction itself were in practice suppressed for a considerable period, whatever the theoretical protections the Bill of Rights may offer for parliamentarians’ freedom of speech.

Super-injunctions are more commonly sought by private individuals seeking to prevent press reports about them and their private lives. In those situations, the use of parliamentary privilege to frustrate a court order might be more difficult to justify. The super-injunction here clearly poses more profound constitutional problems, as it stymied the ability of parliamentarians to hold the executive to account for serious administrative failures and for substantial public expenditure. It is therefore perhaps timely that the House of Commons Procedure Committee is conducting a review of the sub judice rule. Its interaction with super-injunctions will be an important consideration when Parliament reflects on how best to strike the balance between comity and accountability.

Context matters, especially for the ‘political’ constitution

There is, if possible, a further consideration which makes this whole case – and in particular the decision not to involve Parliament in any form – even more constitutionally exceptional. In July 2024, a general election was held. Some polls gave Labour a comfortable lead, others less so. One would be hard-pressed not to think that this further governmental debacle in relation to the (already heavily criticised) retreat from Afghanistan, as well as the financial consequences associated with the emergency settlement in the UK of the people on the leaked list, could have had a significant impact on the electoral chances of the incumbent government. And while Chamberlain J did not specifically mention the upcoming general election in his May 2024 decision to lift the super-injunction, it would not be surprising if the consideration of giving the electorate a chance to hold the government to account at the polls constituted a further reason for him to do so.

After all, this is how we are told the ‘political’ constitution works: government and Parliament make decisions, and every five years (at the latest) the public can hold them to account at the polls. But that implies, of course, that the public is aware of what the government is actually doing. Once again: while Chamberlain J seems acutely aware of this throughout this saga, the fact that the CoA does not seem to give any weight to this foundational tenet of our constitutional system is astonishing.

Even more so from a political constitutionalist lens, context matters: if a super-injunction was granted at the beginning of the life of a new Parliament, an argument (however convincing) could be made that even if the injunction stayed in place some time, this would still leave plenty of time for the electorate to exercise their popular sovereignty at the next elections. But to grant and maintain such an unprecedented* injunction towards the end of the life of a Parliament, and especially throughout a momentous general election, appears nearly impossible to reconcile with a political constitutionalist perspective.

The Afghan leak super-injunction, therefore, directly violated not just one of the two paramount constitutional functions that Parliament is tasked with exercising in our system of government (holding the government to account), but infringed upon the very legitimacy which underpins and justifies the exercise of this function. As such, this case perfectly encapsulates why several commentators, as well as judges, are uneasy about such a derogation from the basic principle of open justice: because an interim non-disclosure order, which by definition should be in place for as little time as possible, could in practice inhibit mechanisms of democratic accountability for nearly two years.

Conclusion

None of this, to be sure, is meant to minimise the very hard choice that  Knowles J was presented with on a quiet day in August 2023. What followed, however, was nothing short of remarkable. Chamberlain J comes out of these proceedings as a constitutional Sisyphus: for months he is pushing up the hill a constitutional boulder not of his making (the unprecedented* super-injunction) and nearly reaches the summit (his order to discharge it in May 2024), only to be sent back to the bottom by the hasty reasoning of the Court of Appeal in July 2024.

Super-injunctions granted to the sitting government are very hard to reconcile with parliamentary democracy. Their justifiability is arguably a matter of two variables: the subject matter protected by the super-injunction, and the length of their duration. A super-injunction which is only granted to the government for a week on a matter which does not require the sudden expenditure of large sums of public money is therefore not nearly as constitutionally problematic as the one under examination here.

Ultimately, several constitutional principles were sacrificed for nearly two years: the principles of open justice, parliamentary sovereignty and accountability, the rule of law, and parliamentary democracy itself. Perhaps what is most extraordinary about this whole case is the fact that it is not the government that is primarily to blame for this bonfire of constitutional principles, unlike the last decade or so.

* I have noted, throughout this post, that this whole situation is ‘unprecedented’. But the nature of super-injunctions is such that we cannot really be sure whether this is the case or not. We – members of the public, lawyers, scholars – have no way of knowing, if not for occasional (and not widely known) questions by parliamentarians. This is why the Report by the Committee on Super-Injunctions suggested the creation of an appropriate ‘data collection system’ to keep a public record of all interim non-disclosure orders (including super-injunctions) made by courts.

My utmost gratitude goes to Anurag Deb, Lewis Graham, Mark Elliott, Paul Scott, Gabriel Tan, Stuart Wallace, Se-shauna Wheatle, Paul Wragg, and to a colleague who wishes to remain anonymous (without the need for a court order) for very helpful discussions, incisive comments and edits on previous drafts of this post.

Paolo Sandro is Associate Professor of Public Law and Legal Theory at the University of Leeds. He is currently an Academic Parliamentary Fellow in the Parliament, Public Administration and Constitution Hub in the House of Commons.

Suggested citation: P. Sandro, ‘Parliamentary democracy under the veil of secrecy: the Afghan leak super-injunction case as a bonfire of constitutional principles’, U.K. Const. L. Blog (23rd July 2025) (available at https://ukconstitutionallaw.org/))