The Attorney General for Northern Ireland asked whether a revised code could treat adults lacking mental capacity as consenting to their care through positive wishes and feelings. The Supreme Court overruled Cheshire West’s ‘acid test’, restored a multifactorial approach and held the Code lawful.
Background
The Attorney General for Northern Ireland referred a devolution issue under paragraph 34 of Schedule 10 to the Northern Ireland Act 1998. The reference concerned the meaning of ‘deprivation of liberty’ under article 5(1) ECHR for adults aged 16 and over. These are people who lack mental capacity under domestic law to decide on their residence and care, and who live in community settings that amount, or may amount, to confinement.
Law Teacher writing support
Writing a law essay?
We choose a writer who is an expert in your area of law and qualified to at least the academic level you select. Most of our writers hold a master’s degree or PhD, have teaching experience, or both.
Send us your question and, if you wish, any marking criteria. We’ll write an essay to your instructions.
The question arises in domestic law as well as under the Convention. Section 306(1) of the Mental Capacity Act (Northern Ireland) 2016 and section 64(5) of the Mental Capacity Act 2005 both define deprivation of liberty by direct reference to article 5(1).
Following HL v United Kingdom, the UK legislatures introduced authorisation schemes for incapacitated people deprived of liberty. In Surrey County Council v P; Cheshire West and Chester Council v P [2014] UKSC 19 (‘Cheshire West’), a majority of the Supreme Court held that the ‘acid test’ was whether the individual was under ‘continuous supervision and control’ and ‘not free to leave’. Compliance, lack of objection, relative normality and purpose were held irrelevant. On the conventional understanding of that case, lack of mental capacity equated to lack of ‘valid consent’ under the subjective element of the Storck v Germany test.
The court noted that over 400,000 people across the United Kingdom are estimated to be deprived of liberty on that approach. DOLS referrals in England rose from about 13,700 in the year before Cheshire West to 332,455 in 2023/24, with a backlog of 123,790 cases. The Law Commission estimated that full compliance would cost £2.2 billion per year.
The Minister of Health for Northern Ireland proposed to issue a Revised Code under section 288(4) of the MCA 2016. Paragraph 2.12 of the draft provided that a person lacking capacity may be regarded as consenting where they actively express positive wishes and feelings about their care arrangements, going beyond mere non-objection. Where such valid consent is given, the statutory authorisation procedures would not apply.
The referred question and the parties’ positions
The referred question was:
Whether the proposed exercise of the Minister of Health’s powers under section 288(4) of the MCA 2016 to revise the Deprivation of Liberty Safeguards Code of Practice issued under section 288 of that Act would be invalid by reason of section 24 of the Northern Ireland Act 1998 as incompatible with article 5 of the Convention.
The Attorney General
The Attorney contended that Cheshire West went beyond the Strasbourg jurisprudence on the subjective element. She said it subjected people who are positively happy with their arrangements to intrusive and costly state processes. She invited the court to depart from Cheshire West so far as necessary to permit the Revised Code.
The interveners
- The Lord Advocate argued that valid consent could, with safeguards, be given by an individual without legal capacity. She also argued that Cheshire West did not decide otherwise.
- The Secretary of State for Health and Social Care argued that the subjective element cannot be considered in isolation from the objective element. He submitted that the law took a wrong turn in Cheshire West by substituting an acid test for a multifactorial analysis, and invited the court to depart from it.
- Mind, Mencap and the National Autistic Society urged the court not to determine the reference in the abstract. They argued that people lacking decision-specific capacity cannot give valid consent, and that a ‘positive attitude’ test is vague and unworkable. They also warned that it would strip vulnerable people of article 5(4) safeguards.
- The Mental Welfare Commission for Scotland supported giving significant weight to expressed wishes and feelings.
- The Official Solicitor warned of the risks of determining the point without a concrete factual matrix, and of loss of access to the courts.
The issues
- What test governs prospective review of a ministerial code under section 24 of the Northern Ireland Act 1998?
- Can the subjective element be addressed in isolation from the objective element?
- Did the majority in Cheshire West misinterpret the Strasbourg jurisprudence?
- If so, should the court depart from it under the 1966 Practice Statement?
- Is the Revised Code compatible with article 5?
The court’s reasoning
The standard of review
The court doubted that a code of practice is subordinate legislation. It held, however, that nothing turned on this, because issuing the code would in any event be an ‘act’ of the Minister.
The highly deferential ‘all or almost all cases’ test applies to devolved primary legislation. It was inapplicable here because the case concerned ministerial discretion, not legislative competence. The court instead adopted the test in R (A) v Secretary of State for the Home Department [2021] UKSC 37:
does the policy in question authorise or approve unlawful conduct by those to whom it is directed?
Transposed to this context, the question was whether the Revised Code authorises or approves unlawful conduct, in the form of a breach of Convention rights, in a significant number of cases. If Cheshire West stood, the Code would fail that test. The correctness of Cheshire West was therefore unavoidable.
Scope of the reference
The court agreed with the Secretary of State that all three Storck elements must be considered together. Valid consent cannot be assessed in isolation, because the starting point must be to identify to what an individual is or is not consenting. Building on a possibly wrong approach to the objective element would risk greater divergence from Strasbourg.
Precedent and the 1966 Practice Statement
The court accepted that departure from its own precedent is exercised ‘rarely and sparingly’. No one suggested that Strasbourg law had changed since 2014. Several factors nonetheless justified departure:
- the direct statutory link between domestic law and article 5, which makes divergence a source of considerable confusion;
- the absence of any route for public authorities to obtain correction of an overbroad interpretation directly from Strasbourg, which places a particular responsibility on the Supreme Court;
- the serious administrative difficulties and unjustified cost generated by Cheshire West;
- repeated attempts by experienced judges to distinguish it, which showed that the decision was impeding the proper development of the law.
The Strasbourg approach: a multifactorial assessment
The court reviewed the Strasbourg case law, including Engel, Guzzardi, Storck, Stanev, HL, HM v Switzerland, Nielsen, Shtukaturov, DD v Lithuania, Mihailovs, Stankov, Austin, Creanga and Ilias and Ahmed.
It held that the starting point is the individual’s concrete situation. Account must be taken of the type, duration, effects and manner of implementation of the measure. The difference between deprivation of liberty and a mere restriction is one of degree or intensity, and the touchstone is closeness to the paradigm of confinement in a prison cell. In marginal cases, purpose is a relevant, though not decisive, factor; Creanga was confined to its police-station context.
The subjective element and valid consent
The central holding was that valid consent is an autonomous Convention concept. It is distinct both from domestic legal capacity and from the free and fully informed consent required to waive Convention rights. The court distilled this principle:
An individual without legal capacity, but who is conscious of their environment and has a basic understanding of their living circumstances in a secure care environment, so that they can in some suitable way express their view about their situation, who manifests their acceptance of that situation, should have their opinion respected when an assessment is made whether they are suffering a deprivation of liberty under article 5. Their subjective attitude, as so expressed, carries significant, indeed usually decisive weight, according to the criteria set out in Storck.
Conversely, a manifested non-acceptance will usually lead to a finding of deprivation where the objective circumstances indicate detention. Mihailovs showed that tacit consent can negative the subjective element. The court treated Stankov, para 90, as concerned with loss of victim status, not with the definition of deprivation of liberty.
The court grounded this approach in respect for human dignity, drawing on Fernandes de Oliveira v Portugal and Stanev, para 153. It held that Baroness Hale’s policy concern was overstated, because individuals in care remain protected by articles 2, 3 and 8 and by other legal regimes.
Why Cheshire West was wrong
The court identified six errors.
- An insufficient test. The acid test may be necessary but is not sufficient to establish a deprivation of liberty. The court stated:
Cheshire West creates a bright-line test which is too crude in its application and which leads to an over-extensive interpretation of deprivation of liberty on the basis of a misplaced policy concern that safeguards in relation to vulnerable people have to be provided pursuant to article 5. In our view, this policy-driven approach puts the cart before the horse.
It added that the majority:
wrongly isolated a small sub-set of factors and elevated it to a universal test that ignores all other potentially relevant factors.
- Compliance and objection. These are relevant to the objective element through the ‘effects and manner of implementation’ of the measure. Objection produces conflict and coercion, whereas contentment points away from deprivation. Mere acquiescence, especially under sedation, may carry little weight.
- Setting and normality. These are relevant factors. Living in one’s own home in accordance with one’s wishes makes a deprivation less likely, so the restrictions would need to be more severe to amount to one.
- Innate limitations. The acid test ignores limitations arising from a person’s own physical or mental condition. Endorsing Lieven J in Re SM and adopting Professor Eldergill’s observation that ‘Mere inability is not lack of freedom’, the court held that people who are unconscious, minimally conscious, catatonic or so profoundly disabled that they cannot conceive of leaving are not deprived of anything. It stressed that ‘Liberty means the same thing for everyone’ and that this is not discrimination.
- Purpose. The majority was wrong to discount the potential relevance of the purpose of the measures.
- Capacity and consent. The majority equated lack of legal capacity with lack of valid consent:
Lack of valid consent as the subjective element of deprivation of liberty is an autonomous Strasbourg concept. It is separate from and not tied to the question of legal capacity in a domestic law context.
The court also criticised the majority’s reasoning on particular points. Baroness Hale’s para 55 got the analysis ‘the wrong way round’: Strasbourg treats incapacity to express any view as pointing away from, not towards, the subjective element. Lord Neuberger confused the criteria for identifying a deprivation of liberty with its justification and with waiver.
The Charities’ objections
The court accepted that there are real evidential difficulties. It nonetheless rejected the argument that, because eliciting wishes is difficult in some cases, wishes should be disregarded in all cases. That approach was:
contrary to the fundamental principle underlying the Convention that individuals should be treated with due regard to their dignity as persons and that their own views should be accorded respect
It set out a safeguard:
If the individual is capable of expressing a view and there is serious doubt about their attitude, no inference should be drawn.
Application to the Cheshire West individuals
By way of guidance, the court considered that MIG and MEG were not deprived of liberty. Both gave sufficient indications of contentment to amount to valid consent, and the objective element was absent. In MEG’s case, the medication was not administered to disable her from forming a view, and restraint was protective rather than punitive. The court reserved its opinion on P, but expressed sympathy with Munby LJ’s view that he was not deprived of liberty.
Practical significance
The decision overrules Cheshire West and replaces the bright-line acid test with the Strasbourg multifactorial assessment. A person’s expressed contentment, compliance, the normality of their setting and the purpose of the measures are restored as relevant considerations. The decision is likely to reduce substantially the number of people treated as deprived of liberty across all UK jurisdictions, given the statutory link to article 5 in the MCA 2005 and the MCA 2016 and its implications for Scotland. This should free resources for cases of genuine confinement.
The court cautioned that the draft Code ‘has not been fully worked through’ and will require revision in light of the judgment. In particular, the draft’s statement that compliance, normality and purpose are irrelevant now conflicts with the court’s reasoning. The court also noted that, as a matter of logic, the approach potentially affects all settings, including hospitals, care homes, supported living and private homes. It stressed that careful consideration is needed to guard against acquiescence, masking and the difficulties of assessing the wishes of individuals with very varied communication abilities.
Verdict: The Supreme Court unanimously (Lord Sales and Lady Simler, with whom Lord Reed, Lord Hodge, Lord Lloyd-Jones, Lord Stephens and Lady Rose agreed) applied the 1966 Practice Statement, declined to follow Cheshire West and overruled it, holding that the ‘acid test’ had never been adopted by the European court and was wrong in principle. The referred question was answered in the negative: the Minister would not be acting incompatibly with article 5 in issuing the proposed Revised Code, which neither authorises nor approves unlawful conduct, and it is within the Minister’s competence under section 24 of the Northern Ireland Act 1998 to issue it, although the draft will require revision in the light of the judgment.
