Alison Young: Declaratory Orders and Constitutional Guardrails

The recent decision of the Supreme Court in Craig v HM Advocate (for the Government of the United States of America) affirms the core constitutional expectation that the executive will comply with declaratory orders. On its face, this appears to be quite unassuming. Yet it raises fundamental issues of inter-institutional comity and the importance of constitutional guardrails and safeguards. 

Craig had been accused of taking part in a fraudulent scheme, providing false information on twitter to reduce the share prices of US-based companies, only to buy the stock cheap and sell it at a higher price at a later date. He faced an extradition order from the US. Craig sought to defend the extradition order. He argued that that it was forum barred and, later, that extradition would breach his Convention rights. The forum bar provisions were contained in the Crime and Courts Act 2013, which provided a set of criteria to use to assess whether, given that the activity of the alleged crime took place in the UK, the UK would be the most appropriate forum for a trial. Problems arose, as the legislation implementing the forum bar – the Crime and Courts Act 2013 – had not been brought into force in Scotland. 

Craig’s extradition hearing was suspended in order to enable him to challenge the decision not to bring the forum bar into force in Scotland. He was successful in his challenge. The Lord Ordinary, Lord Malcolm, concluded in the Outer House of the Court of Session that it was unlawful for the Government not to have used its powers to bring the forum bar into force in Scotland when it had been brought into force in England and Wales. The commencement provision found in section 61 of the 2013 Act, whilst permitting some provisions of the legislation to come into force at different times across the UK, required the forum bar provisions to be commenced at the same time across the UK as a whole. Consequently, Lord Malcolm ‘found and declared that in its continuing failure’ to bring the forum bar provisions into force in Scotland, that the UK Government was ‘acting unlawfully and contrary to its duties under section 61’ of the 2013 Act. No appeal was lodged against this decision. 

Having obtained this decision, Craig continued with his extradition hearing. He argued that the extradition order breached his Convention rights, namely his right to family life. In particular, he argued that the breach of his right to family life could not be justified as the restriction on his rights from the extradition order was not ‘in accordance with the law’ given the continued failure of the UK government to implement the forum bar. The Lord Advocate, however, argued that the failure to implement the forum bar was a mere procedural error that would make no difference to the outcome. Whether the extradition order breached convention rights could still be determined, with the courts also able to take account of the non-implementation of the forum bar when making this assessment. After a series of appeals, the matter was appealed to the Supreme Court on a devolution issue. Craig succeeded. The breach of his Convention rights was not ‘in accordance with the law’ and, consequently, the extradition order was unlawful.

The importance of the case, however, lies elsewhere: what are the consequences of a declaratory order? No order for specific performance had been granted. So there was no court order requiring the Government to bring the forum bar into force. Nevertheless, a declaratory order had made it clear that it was unlawful for the Government to continue to fail to bring the provision into force in Scotland. Did this mean that there was no need for the UK government to issue a commencement order for the forum bar until an individual had gone to court and succeeded in obtaining specific performance? Lord Reed was clear that this was not the case given the constitutional consequences of a declaratory order. 

There is a clear expectation that the executive will comply  with declaratory orders. Indeed, this expectation underpins the decision of the court to exercise its discretion to issue a declaratory order in the first place. Courts refrain from issuing a coercive order because of the expectation that the executive will follow a declaratory order. This expectation is borne out by practice. The Government has, in the past, accepted that they should normally comply with declaratory orders and can normally be relied upon to do so. Lord Reed went so far as to hope that the submissions made in Craig by the Government were not an indication of a departure from this long-standing approach.

More fundamentally, the expectation that the executive will comply with a declaratory order is ‘one of the core principles of our constitution and is vital to the mutual trust which underpins the relationship between the Government and the courts’ [46]. Declaratory orders work as an effective remedy because of this mutual trust. This mutual trust is evidence of the fact that the UK is a society governed by the rule of law, where the ‘Government can be trusted to comply with court orders without having to be coerced’ [46].

Guardrails and Guardians

The UK constitution seems recently to have been in a perpetual state of change. The post-Brexit constitution is still being formed, with yet more potential reforms to the form and substance of Retained EU law on the horizon. The Conservative Government’s manifesto promise of constitutional change has already seen: the Dissolution and Calling of Parliament Bill; the Independent Review of Administrative Law, a further Government consultation exercise and the Judicial Review and Courts Bill;  and the Government’s Human Rights Act Reform consultation following the Independent Human Rights Act Review. All of this is taking place against a backdrop of a global pandemic and, more recently, war in Ukraine. One can be forgiven for thinking, therefore, that the status of declaratory orders should be anything but an insignificant blip on the constitutional radar. 

However, Lord Reed’s recognition of the importance of mutual trust is as constitutionally significant as these rather more dramatic changes. It has become almost a constitutional hobby horse, post the Brexit referendum, to focus on the extent to which the UK constitution can be described as displaying signs of populism or, worse, autocratic legalism. A blog post is not the place in which to enter into the complex relationship between populism and constitutionalism, or populism and authoritarianism. However, the discussion of mutual trust in Craig raises related issues that may help provide a constitutional health check. 

In their assessment of populism and its impact on democracy, Levitsky and Ziblatt refer to the importance of constitutional guardrails. These are informal norms that underpin and supplement constitutions – codified or otherwise. Like constitutional conventions, constitutional guardrails are widely known and respected, but not legally enforceable. Unlike some conceptions of constitutional conventions, they are norms. They also perform a more fundamental role, sustaining as opposed to merely supplementing constitutional rules. For Levitsky and Ziblatt, the two main guardrails are mutual toleration and institutional forbearance. Institutions exercise mutual forbearance when they exercise self-control, not exercising a legal power in a manner that, whilst legal, undermines the spirit of the law. This generates mutual trust between institutions. 

It can be contrasted with what Tushnet refers to as constitutional hardball – those situations when institutions play within the rules, but do so in a manner that is in tension with, or even undermines, pre-existing constitutional understandings as to how a legal power should be exercised. Taken to its extreme, playing constitutional hardball can lead to what Kim Lane Scheppele refers to as autocratic legalism, using the legal and constitutional rules to change the constitution in a manner designed to undermine the values underpinning constitutionalism. 

This is not to argue that Craig shows that the UK government is playing constitutional hardball. If anything, it shows that the need for mutual trust and institutional forbearance may give rise to difficult conflicts. Whilst the Outer House had made it clear that it was for the UK government to decide when to commence the forum bar provisions, this being applied across the UK as a whole, it was also clear that there were misgivings about the forum bar in Scotland. Mutual trust between Westminster and the Scottish government may have pushed in one direction, whereas mutual trust between the courts and the executive pushed in another. However, clear legal obligations provide a hard-edge. 

In this sense, the court is playing its role as constitutional guardian, but doing so in a manner that recognises and respects the role of other institutions. There is no need to force the executive to act through issuing a court order when there is mutual trust. Courts can exercise institutional forbearance and trust the government to comply with a legal rule without the need for specific performance or a court order. When that trust breaks down, then courts can use their legal powers to ensure the executive adheres to the rule of law. To do so is not to play constitutional hardball. It’s a measure of constitutional counter-balancing. When the constitutional guardrails start to bend and mutual trust is not adhered to, then a constitutional guardian needs to bend the guardrail back into shape, enforcing a legal rule and, when needed, issuing a court order. This was not needed in this instance, where the UK Government did enact a commencement order, albeit two years after the declaratory order and at a time that was too late to apply to Craig’s situation. But, even if the court had needed, at a later date in a different application, to issue an order for specific performance, this would not have been an example of constitutional hardball undermining mutual trust. It would have been a response to an undermining of mutual trust. That does not give rise to autocratic legalism, especially when it is used to shore up a constitutional norm – here the rule of law. 

Mutual trust or mutual mistrust?

The recognition of mutual trust underpinning the relationship between the Government and the court is timely. This is the second time within a year that the Supreme Court has been called upon to deal with situations in which the government has, ‘exceptionally’ in the words of Lord Reed, failed to comply with a court order. In Majera, the government ignored a court order, because it was of the opinion that it was unlawful, therefore void, and so could be ignored. The Supreme Court made it clear, however, that the government had to obey a court order. If the government believed the order was wrong it could appeal the court order. 

However, arguably there is a deeper sense of a lack of trust, evidenced by the Government’s response to the Independent Review of Administrative Law and the Independent Human Rights Act Review. Both refer to examples of cases where the Government thinks the courts went too far, playing constitutional hardball in a manner that failed to respect the pre-constitutional understandings determining the proper role of the court. These examples, as recognised in part in the Government’s response to the Independent Human Rights Act Review, are as exceptional as the situations in which the government fails to adhere to a declaratory order, or to respect a court order. For the most part, they occurred at times of constitutional transition, or some would even argue a constitutional crisis, when pre-constitutional understandings are in flux. 

Pre-constitutional understandings are constitutionally contestable. They play an even more fundamental role in the UK constitution, where there are fewer legally enforceable rules determining the scope of executive and judicial power, with greater reliance on inter-institutional respect and mutual tolerance. If these constitutional guardrails are to continue to perform their job, then it is time for them to be discussed openly and widely, and not just on the pages of specialist constitutional law blogs.  To achieve this, there is a need for broader public education on constitutional issues, and wider public debate about current constitutional reforms. Otherwise important constitutional change may also run the risk of becoming another blip on the radar. More fundamentally, constitutional change without public engagement may make it harder to distinguish between a legitimate change of contestable constitutional principles from autocratic constitutional change that merely purports to represent the views of the people. 

With thanks to Mike Gordon and Hayley J Hooper for their comments on an earlier version. 

Alison Young is the Sir David Williams Professor of Public Law at the University of Cambridge and a Fellow of Robinson College.

(Suggested citation: A. Young, ‘Declaratory Orders and Constitutional Guardrails’, U.K. Const. L. Blog (10th March 2022) (available at