In A Reference by the Attorney General for Northern Ireland of a devolution issue under paragraph 34 of Schedule 10 to the Northern Ireland Act 1998 [2026] UKSC 16, a seven member Panel of the Supreme Court has unanimously overruled the 2014 majority decision in Cheshire West. The judgment is long and detailed. A summary by the Supreme Court can be found here, and a more detailed summary prepared by members of 39 Essex Chambers will be forthcoming soon. In the meantime, the key reasoning of the Supreme Court is summarised at paragraphs 52 to 55 as follows:
52. The question on this reference is directed at and invites consideration only of the subjective element (concerning lack of valid consent to confinement) in article 5. However, the fundamental question is whether the individual has been “deprived of his liberty” as that expression is used in article 5. We agree with the Secretary of State that this requires consideration of all three elements of the deprivation of liberty test taken together. It is not possible to consider the concept of absence of valid consent in isolation. As we explain below (para 126), “valid consent” for these purposes is distinct from consent to an infringement of Convention rights as required for a waiver of rights. Valid consent in the present context is an aspect of the assessment whether the relevant parts of article 5 predicated on there being a deprivation of liberty are applicable at all. In that context the starting point must be to identify to what an individual is or is not consenting, and that inevitably requires consideration of the objective element, regarding the circumstances of the confinement. To do otherwise risks developing domestic law by building on what may be a wrong approach to the objective element. This would be contrary to sections 306 of the MCA 2016 and section 64(5) of the MCA 2005, which define deprivation of liberty by reference to its meaning under article 5 and, if the Attorney and the Secretary of State are correct in their submissions, could lead to an even greater departure from the authoritative interpretation of that concept given by the European court. It follows, in our view, that the reference requires us to consider all three elements of the article 5 test.
53 For the reasons explained in detail below, we would answer the referred question in summary as follows:
(i) The starting point in assessing whether someone has been deprived of liberty within the meaning of article 5 is the specific situation of the individual concerned, and the assessment is multifactorial, with account taken of a whole range of factors including the type, duration, effects and manner of implementation of the measure in question. This was the approach set out in Engel v The Netherlands (1979–80) 1 EHRR 647 (“Engel”) and Guzzardi v Italy (1981) 3 EHRR 333 (“Guzzardi”) and it has been consistently applied by the European court since then, and in many different contexts, including the one with which we are here concerned. The judgments of the European court show that no single factor is determinative.
(ii) In addition to the objective element of confinement in a restricted space for a significant period, an individual will only be considered to be deprived of liberty if, as an additional subjective element, they have not validly consented to the confinement in question (Storck). Valid consent is therefore a powerful factor in the assessment. It is an autonomous concept and not to be equated with the concepts of consent for the purpose of waiver of rights under the Convention or of legal capacity in domestic law. The fact that an individual lacks legal capacity to decide on their living and care arrangements does not necessarily mean that they are de facto unable to understand and consent to those arrangements in a manner that prevents those arrangements from becoming a deprivation of liberty (see Stanev, HL and the other judgments of the European court considered below). On the contrary, an individual without legal capacity under domestic law, but who is conscious of their environment and has a basic understanding of their living circumstances so that they can express their view about their situation, who manifests their acceptance of the situation they are in, should have their opinion respected when an assessment is made of whether they are deprived of liberty under article 5.
(iii) Although the objective and subjective elements of deprivation of liberty are often considered sequentially by the European court in its assessment of an applicant’s specific situation, there is an overlap. The requirement to take account of the “type” and “effects and manner of implementation” of the measure in question means that the assessment of the objective element can take account of the specific context and circumstances of restrictive measures that are different from the paradigm of confinement in a cell.
(iv) The approach of the European court means that the effect of restrictions on an individual, including their compliance and the lack of objection if they are capable of objecting or giving tacit agreement, is relevant in assessing the objective element of confinement. The relative normality of the placement is also a relevant factor in this assessment. Similarly, in situations that are far from the paradigm of confinement in a cell, the purpose for which a measure has been taken is a factor to be considered.
(v) The European court has recognised that the process of assessing whether there has been a deprivation of liberty is no easy task in some contexts and may give rise to difficulties, especially in borderline or marginal cases. Equally, it may sometimes be difficult to ascertain the true feelings or preferences of vulnerable individuals who do not have mental capacity to decide on their living arrangements. The approach should be practical and realistic. Where there is serious doubt, no inference of valid consent should be drawn.
(vi) It follows from the analysis of the Strasbourg jurisprudence in section 8 below that, in setting out the acid test, the majority decision in Cheshire West departed from the longstanding multifactorial approach to determining whether a person is deprived of liberty within the meaning of article 5 and is wrong (as we explain in section 9 below).
(vii) For the reasons given in section 3 above, we have concluded that the 1966 Practice Statement should be applied and that Cheshire West should be overruled.
(viii) The Minister would not be acting incompatibly with article 5 in issuing the Revised Code and it is therefore within competence for him to proceed to issue it.
54. In reaching these conclusions we acknowledge the fact that this reference concerns a group of people who lack the mental capacity under domestic law to consent to their care arrangements. Moreover, as the evidence of the Charities makes clear, people with mental health problems or learning disabilities and autistic people have a wide range of abilities and care needs. Some can communicate through speech, some are non-verbal and use “Makaton” or assistive technology, and some communicate through sounds or gestures. Some can understand the different possible options for their care but cannot weigh them up to make a choice; others are unable to understand what their support needs are, let alone the different ways and means in which they could be met. There is also a wide spectrum of care arrangements in the community for disabled people who lack capacity to decide on their residence and care. Some live in their own homes which they own or rent, or with parents or foster parents in the family home. Some live in supported living placements (with a tenancy agreement signed on their behalf) which have communal areas and shared staff, or they may live in a care home with other residents. There is similarly a wide spectrum of care arrangements for those who are detained in institutional settings and hospitals, ranging from those who are detained in “open” hospitals to those who are confined to a small space under close supervision. Care plans, even for those who live in their own homes in the community, can involve a wide range of restrictive measures that are considered to be in the individual’s interests (for example, the use of sedative medicines, physical restraint devices, electronic and other monitoring) but which have the potential to be overused, misused or abused. Put another way, even a domestic setting can involve a deprivation of liberty
55. We do not assume that because an individual is cared for in a community setting, their care arrangements are necessarily less restrictive than in a psychiatric hospital, nor that an individual subject to restrictive care arrangements would inevitably object to them or indicate that they were not happy to live in the particular setting. Nor do we under-estimate the difficulty of identifying whether a disabled or mentally impaired individual subject to such measures is agreeing to them, objecting to them, or otherwise has a “positive attitude to the care arrangements” (see para 2.12 of the Revised Code), as demonstrated by the evidence submitted by the Charities. These points will require careful consideration in finalising the Revised Code and in the further guidelines we understand will be produced. Moreover, although the Attorney’s written case suggests that the Revised Code would not apply to mental health hospitals in Northern Ireland, the approach we consider correct in relation to the subjective element of valid consent would, as a matter of logic, potentially affect all settings, including hospitals, care homes, children’s homes, hostels, supported living arrangements and people’s own private homes. This too will require further consideration to ensure appropriate provision is made.
It should be noted that there is no ‘grace period’ for implementation of the judgment: i.e. the approach in Cheshire West must not be followed with immediate effect, and all the (many) guidance documents which refer to it need to be read in that light. It is likely that official guidance will be forthcoming soon.
