We are all bankers now

Before long, significant changes will have been made to the law that covers children and many incapable or vulnerable adults. It is still unclear, however, what those changes will mean.

Yet the law must always be clear. The European Court of Human Rights (ECtHR) said as much in 2004: “An important ingredient of lawfulness is that all law must be sufficiently precise to allow the citizen – if need be with appropriate advice – to foresee, to a degree that is reasonable in the circumstances, the consequences which a given action might entail.” (HL v United Kingdom) In fact, this principle goes back much further; at least to Sunday Times v United Kingdom in 1979 and possibly to the ancient Chinese Fa legalism of the eighth to third centuries BCE.

Hello DoLS

The HL case was about the common law doctrine of necessity and whether it was sufficiently clear to permit incapable people to be deprived of liberty in their own best interests. The ECtHR said it was not and that as a result, it breached Article 5(1) of the ECHR.

The Government’s response was to introduce the Deprivation of Liberty Safeguards (DoLS), which will apply wherever an incapable person is in a NHS hospital or premises registered under the Care Standards Act. For such a person to be deprived of liberty, permission will have to be obtained, either from a PCT or a local authority. If the Court of Protection is to be involved in the case, it will be subsequently, as an appellate court.

One problem with the DoLS is that even now, some time after they came into effect, no one really knows what it means to be deprived of liberty. And it seems that the best guidance is as likely to be found in anti-terrorism cases as in those decided under health care law. (See: David Hewitt, New perspectives on the Mental Health Act, Solicitors Journal, 18 November 2008) Practitioners – and patients – surely deserve better than that.

Another problem is the attitude of the House of Lords, which recently held that in deciding whether something constitutes a deprivation of liberty, consideration may be given to the purpose it seeks to achieve. (Austin v Metropolitan Police Commissioner, 2009) Their Lordships drew on HM v Switzerland (2002), in which the ECtHR held that where an elderly woman was admitted to a nursing home, Article 5 of the ECHR would not be engaged, because this was a “responsible measure” taken in her own best interests and could not, therefore, amount to a deprivation of liberty. The decision in HM had long been considered a rogue one, and In HL, a differently-constituted ECtHR took pains to distinguish it. Now, however, its spectre has returned, and in a way that might haunt the DoLS. (See: David Hewitt, Whose liberty? Solicitors Journal, 13 February 2009)

Under the DoLS, a deprivation of liberty will only be permitted – and will only, therefore, be lawful – if it is in the best interests of the incapable person concerned. But Austin suggests that a measure will never be a deprivation of liberty where it has a benevolent purpose. If that were so, no patient who appeared to fall within the DoLS would ever be deprived of liberty, and as a result, permission would be unnecessary where the DoLS appeared to demand it most. No formal comment has been made about the effect of Austin, but the Department of Health has let it be known that as far as the DoLS are concerned, it is business as usual.

No secrets

There is a widespread view – even unanimity, perhaps – that we should strive to protect vulnerable people. The problem is that no one can seem to agree on what ‘vulnerable’ means. (See: David Hewitt, What is vulnerable? Solicitors Journal, 14 April 2009)

The word is used in the No secrets guidance, which was published in 2000 and is the centrepiece of the adult protection framework. The guidance is, however, being revised, and there is no guarantee that vulnerability will continue to have the resonance it does now. And the ‘No secrets’ definition is not the only one to hand: the Youth Justice and Criminal Evidence Act 1999, the Care Standards Act 2000 and the Safeguarding Vulnerable Groups Act 2006 all have to do with the vulnerable, but each of them defines the term differently.

The High Court, meanwhile has said that it too will offer a remedy to the vulnerable. (Re SA: vulnerable adult with capacity: marriage, 2006) The Court’s motive might be to create new work streams, given that it has lost much if not all of its jurisdiction over incapable people to the Court of Protection. Its conception of what it means to be vulnerable is certainly the broadest of them all.

Parental consent

Hitherto, someone with parental responsibility has been able to consent to medical treatment for a child, even where the child himself or herself is capable of giving consent, and even where he or she has refused to do so. Now, however, we are told that sometimes, it will not be sufficient to rely upon parental consent. (See: David Hewitt, Too young to decide, Solicitors Journal, 30 September 2008)

This notion is set out for the first time in the revised Code of Practice to the Mental Health Act 1983, where it is said to derive from Nielsen v Denmark (1989). The Code says: “It is difficult to have clear rules about what may fall in the zone, when so much depends on the particular facts of each case … The parameters of the zone will vary from one case to the next: they are determined not only by social norms, but also by the circumstances and dynamics of a specific parent and child or young person.” (Paragraphs 36.9 and 36.12) Furthermore, “Where there is doubt professionals should take legal advice so that account may be taken of the most recent case law.” (Paragraph 36.9)

According to the Code, the factors that should be taken into account include: the nature and invasiveness of the treatment; whether the child is resisting; any conflict between the child and the parent, or between the parents; any discrepancy between the child’s and the parents’ interests; and the mental capacity of one or other parent.

But another factor, which might itself render unreliable consent obtained from a parent, is “The general social standards in force at the time concerning the sorts of decisions it is acceptable for parents to make – anything that goes beyond the kind of decisions parents routinely make will be more suspect.” (Paragraph 36.12) And the Code reminds us that a decision might be beyond a parent to make simply because of the nature of the proposed treatment – for example, where it could be considered “particularly invasive or controversial” (Paragraph 36.14) But does this guidance clarify or simply obscure the position?

What, for example, are the general social standards in force at the time concerning the sorts of decisions it is acceptable for parents to make; and how should we go about identifying them? Are practitioners expected to take the Guardian and the Daily Mail, and to watch both Newsnight and Jeremy Kyle? And is a code of practice really the right place for such a far-reaching proclamation? If parental consent is no longer to be a sound basis for medical treatment given to a child, why doesn’t the Government say so explicitly, in legislation?

In cases of doubt, professionals are advised to take legal advice, “so that account may be taken of the most recent case law.” This is because the Government no longer regards the Nielsen decision as reliable. The most recent case law, however, supports Nielsen: the decision was another of the golden oldies revived by the House of Lords in Austin. Does that mean, therefore, that on the basis of the Code of Practice, parental consent is still sovereign after all?

And finally: is a decision about cancer surgery really – and by reason only of its magnitude – beyond the remit of a child’s parent? If it is, can we say with any degree of certainty that such a state-of-affairs is itself consistent with the general social standards concerning the sorts of decisions it is acceptable for parents to make?

Comment

These changes are worryingly vague, both in their implications and in their own terms. But if each change is taken at face value, it produces some results that are startlingly clear.

If, ultimately, a judge is to decide whether a child can have medical treatment, a vulnerable adult be protected or incapable person deprived of liberty, a great measure of professional discretion will have to be surrendered. And if the judge that makes such decisions sits in the Court of Protection or the Family Division, and not the High Court, his or her remit will be wider and the yardstick will be merits, not reasonableness.

This will change the practice across wide areas of health and welfare law, so that the judge ceases to be a simple reviewer of public decisions and becomes instead a primary decision-maker. It would be understandable if, in these times of travail, there were to be a loss of faith in professionals. But it’s still a bit of shock to realise that as far as the government is concerned, we are all bankers now.

Whose consent?

Where a child requires medical treatment, the old rule was that the parents could consent on his or her behalf. Now, it seems the old rule is also the new rule.

For the last 20 years or so, it has been accepted that a parent can consent to medical treatment for a child. This position is the result of Nielsen v Denmark (1989) 11 EHRR 175 and it will obtain, even if the child is competent and has refused consent. In Nielsen, a boy who did not want to go into a closed psychiatric ward was admitted there with the consent of his mother.

Late last year, the Department of Health cast doubt upon this position, at least in the case of competent children. In the revised Code of Practice to the Mental Health Act 1983, it noted that Nielsen was decided before the Human Rights Act 1998 came into effect and “the trend in recent cases is to reflect greater autonomy for under 18s in law”, and concluded: “it is not wise to rely on the consent of a person with parental responsibility to treat a young person who refuses in these circumstances.” (Department of Health, 2008, Mental Health Act 1983 Code of Practice, paragraphs 36.33 and 36.43)

This conclusion caused dismay among practitioners in child and adolescent mental health. More recently, however, and despite what the Department of Health said, the House of Lords has suggested that Nielsen remains good law.

In Austin v Commissioner of Police for the Metropolis [2009] UKHL 5, their Lordships found, perhaps contentiously, that a woman who took part in the May Day protests of 2001 had not been deprived of liberty when, along with several thousand other people, she was corralled into Oxford Circus by the police and detained there for a number of hours. They said that the purpose of these measures – crowd control – was a permissible one and “in the interests of the community”, so that Article 5 of the ECHR, which enshrines the right to liberty, wasn’t even engaged.

In support of their contention that the purpose of the measures was relevant to the question of whether Article 5 was engaged, their Lordships examined Nielsen and found it to be directly relevant to the case before them. They went as far as to cite with approval the following passage of the judgment in that case: “Hospitalisation was decided upon by [the mother] in accordance with expert medical advice. It must be possible for a child like the applicant to be admitted to hospital at the request of the holder of parental rights”.

If, as seems likely, Nielsen remains – or has again become – good law, the concerns of the Department of Health are baseless and any child, even an unwilling one, may be treated with the consent of his parent.

Whose liberty?

Photo: Peter Jordan, PA

The House of Lords recently held that a woman who took part in the May Day protests of 2001, and who was one of several thousand people whom the police confined in Oxford Circus for up to seven hours, was not deprived of liberty. (Austin v Metropolitan Police Commissioner [2009] UKHL 5) The woman lost her case because their Lordships decided that the reason the police did what they did was relevant, and because they accepted that what the police did was proportionate and done in good faith.

Giving the lead judgment, Lord Hope cited a number of cases brought under the European Convention on Human Rights (ECHR), which, he said, established that in deciding whether there is a deprivation of liberty, “Account must be taken of […] the context in which the restriction of liberty occurs” and, therefore, of the purpose of the measures used. (Paragraph 21) Here, the purpose – crowd control – was a permissible one and “in the interests of the community”, so that Article 5 of the ECHR, which enshrines the right to liberty, wasn’t even engaged. This conclusion is likely to be deeply contentious, not least because the authorities do not mention ‘purpose’ at all; rather, they focus on the “type, duration, effects and manner of implementation of the measure in question”. (See, for example: Engel v The Netherlands (1979-80) 1 EHRR 647; Guzzardi v Italy (1981) 3 EHRR 333) These are plainly not the same things.

Lord Hope’s conclusion would appear to owe a great deal to the controversial European Court of Human Rights (ECtHR) decision in HM v Switzerland (2004) 38 EHRR 17. (See: David Hewitt, Whatever Happened to Winterwerp? Solicitors Journal, 10 May 2002, p 436) It did not find favour with Lord Walker, who argued that purpose was irrelevant to the question whether confinement amounted to a deprivation of liberty: “If confinement amounting to deprivation of liberty and personal security is established, good intentions cannot make up for any deficiencies in justification of the confinement”. (Paragraph 44) This, though, was a minority view. The other three members of the Appellate Committee fell in behind Lord Hope and, as she had been in the Court of Appeal - and also, though for slightly different reasons, at first instance - the claimant was unsuccessful.

The possibility that what would otherwise be deprivation of liberty will be prevented from being so by mere good intentions might surprise, and possibly infuriate, the government. Though it could easily have done so, the ECtHR did not reach this conclusion in the ‘Bournewood’ case. There, the Court found that Article 5 was engaged – and had been breached – where Mr L, a man with autism, had been confined to a mental health hospital for three months and prevented from seeing his carers. The Court said this, and not much more, amounted to an assumption of “effective, unqualified control” over the man. (HL v United Kingdom (2004) 40 EHRR 761) Lord Hope did not refer to this aspect of the case.

Because of Bournewood, the government has introduced the Deprivation of Liberty Safeguards (DoLS), which from 1 April 2009 will allow incapable people to be deprived of liberty if certain forms of protection are in place.

It was accepted that the measures used upon Mr L had been in his best interests. If, as the House of Lords has just done, the ECtHR had allowed that fact to prevent Article 5 being engaged, the government would neither have lost the Bournewood case nor been forced to introduce the DoLS. As it is, the DoLS will be relevant where an incapable person is deprived of liberty in his own best interests. Yet if strict observation of best interests will prevent there being a deprivation of liberty, there will be no patient to whom the DoLS apply. Could it be that even before they have come into force, the DoLS are otiose?

And the implications of their Lordships decision go further. If, no matter how constraining they might be, measures taken in the best interests of a patient will never engage Article 5, what is to become of people with mental disorder who are detained under the Mental Health Act? How are they to be protected against unlawful compulsion and what are we to make of the dozens of cases such people have brought, and sometimes won, claiming an impermissible breach of the right to liberty? This baleful question first arose after HM v Switzerland was decided in Strasbourg. It seems it must now be asked again, only this time, much closer to home.

Seconds out ...

The second edition of my book, The Nearest Relative Handbook, is now available.

The book is intended to be a lucid, concise guide the role of the nearest relative and the way it might fall - or be given - to an individual. It should be of use not only to nearest relatives themselves, but also to Approved Mental Health Professionals, Responsible Clinicians and other practitioners; to patients, their families and carers; and to hospital managers, police officers, lawyers, academics and policy-makers.

The Nearest Relative Handbook sets out and explains the complex criteria by which a nearest relative is to be identified and the grounds upon which he or she may be displaced by the court, and it does that not only for adults, but also for children. The book also considers the duty of consultation that some professionals owe to nearest relatives, and other, associated issues, such as Mental Health Review Tribunal rights and information-sharing.

The book has been completely re-written for its second edition, and now includes a wealth of material on the changes made by the Mental Health Act 2007, which have particularly affected the process and grounds for displacement. It also has nearly 50 per cent more practical scenarios.

Review of the first edition were very favourable. They include:

An excellent, comprehensive and thoughtful guide to the rights, powers, and duties of nearest relatives under the Mental Health Act. This is the definitive work on the subject - Phil Fennell, Professor of Law, Cardiff Law School

This intelligent and comprehensive analysis is a welcome addition to the currently scant literature in this area - Journal of Mental Health Law

The rules are comprehensively worked through and the examples clear - Tony Eaton, Solicitor, Brent Community Law Centre

There can hardly be a professional concerned with the Mental Health Act 1983 and its practical application who will not benefit from having this book to hand - New Law Journal

The second edition of The Nearest Relative Handbook is published by Jessica Kingsley Publishers at £17.99. Further details, together with an order form, may be found here

What is a public place?

Where a person who is in a public place appears to be suffering from mental disorder, section 136 of the Mental Health Act allows him to be arrested by a police constable and detained for up to 72 hours in an appropriate place of safety.

It is hard to find definitive statistics on section 136. The Mental Health Act Commission has suggested that between 2002/3 and 2003/4, the power was used 4,450 times, with patients detained either in hospital or in a police station. (MHAC, 2006, In Place of Fear? Eleventh Biennial Report, 2003-2005, paragraph 4.165 and figure 76) The Independent Police Complaints Commission, however, has calculated that in 2005/6, patients detained under section 136 were held in police cells on 11,517 occasions. (Maria Docking et al, September 2008, Police Custody as a “Place of Safety”: Examining the Use of Section 136 of the Mental Health Act 1983, IPCC Research and Statistics Series: Paper 11, pages 10 & 11)

Because the concept of a public place is used in a wide variety of statutes, the cases it produces are rarely collected together, and as a result, it is hard to know quite what the law permits.

The new Reference Guide to the Mental Health Act 1983 says that a ‘public place’ is “any place (whether indoors or outdoors) to which the public have access, whether by right, by explicit or implied permission, on payment, or otherwise.” (Paragraph 30.17). Because of the courts, however, we also know: that the grounds of a bail hostel have been held not to be a public place. (Harriott v DPP [2005] EWHC (Admin) 965) The contrary may be true both of the upper landing and of a communal balcony in a block of flats. (Knox v Anderson (1983) 76 Cr App R 156; Carter v Metropolitan Police Commissioner [1975] 1 WLR 507)The front garden of a private house will not usually be a public place, (R v Edwards (1978) 67 Cr App R 228) even if it is so small that someone standing in it could inflict harm on a pedestrian on an adjacent pavement. (R v Leroy Lloyd Roberts [2003] EWCA Crim 2753) A car park is a public place, at least during licensing hours, (David Lewis v DPP [2004] EWHC (Admin) 3081) and, even if it is said to be for customers only, if it is attached to commercial premises and accessible from a public place. (May v DPP [2005] EWHC (Admin) 1280) Finally, a private driveway shared by neighbouring private properties is not a public place; members of the public who use it do so as visitors rather than as members of the public. (R v Bogdal [2008] EWCA Crim 1)

It is felt that sometimes, where a person with mental disorder is found in a private place, police constables will attempt to entice him into a public place simply so that the section 136 power can be used. The standard texts say that such a tactic would be unlawful, but a brand new case suggests that it is not.

In that case, a police officer was held to have acted lawfully where he arrested a woman for being drunk and disorderly in a public place, having physically escorted her from a garden to a private footpath. The High Court accepted that the constable had been attempting to reach a negotiated conclusion that would have been in the woman’s own interests, and that he was not motivated by the desire to justify an arrest for an offence that could only be committed in a public place. The High Court said he could properly be said to have acted in conformity with generally acceptable standards of conduct. (McMillan v CPS [2008] EWHC (Admin) 1457)

In order to understand the limits of the section 136 power, practitioners will have to look beyond the Mental Health Act, which was responsible for only one of the cases mentioned above. The other cases were brought under statutes as diverse as the Public Order Act 1936, the Prevention of Crime Act 1953, the Criminal Justice Act 1988, the Road Traffic Act 1988 and the Dangerous Dogs Act 1991. Conventional guidance is rarely so diverse as to embrace all these sources.

What is vulnerable?

That is likely to be one of the hot questions of the next few months. It has been posed most recently by the Department of Health, which is reviewing adult protection guidance published in 2000. (Department of Health and Home Office, October 2008, Safeguarding Adults) At the moment, the guidance – and the safeguards to which it gives access - relate solely to ‘vulnerable’ adults. But what does that mean?

Vulnerability
The notion of vulnerability precedes the adult protection guidance. It is used in the Care Standards Act 2000, for example, and also at the heart of the special measures available to certain witnesses in criminal proceedings. (Youth Justice and Criminal Evidence Act 1999, section 16)

The guidance of 2000 is contained in the No secrets document. It says that a vulnerable adult is one who needs community care services because of disability, age or illness; and who cannot take care of himself, or protect himself against significant harm or exploitation.” (Department of Health and Home Office, 2000, No Secrets: Guidance on Developing and Implementing Multi-agency Policies and Procedures to Protect Vulnerable Adults from Abuse) Such a person might expect to be protected by a comprehensive ‘inter-agency’ framework.

There is concern, however, that this framework is more restricted than it should be, and that the problem is one of definition.

The House of Commons Health Committee, for example, says No secrets should not be confined to people requiring community care services, and that it should also apply to old people living in their own homes without professional support and anyone who can take care of themselves. (House of Commons Health Committee, 2004, Elder Abuse, Second Report of the Session 2003-04, Volume 1, HC 111-I, paragraphs 8 & 14)

This echoes the Association of Directors of Adult Social Services (ADASS), which has argued that ‘vulnerability’ “seems to locate the cause of abuse with the victim, rather than placing responsibility with the acts or omissions of others.” (ADASS, 2005, Safeguarding Adults: A National Framework of Standards, page 5)

As the new consultation document notes, there is “a broad belief that the definition does need revision, but no clear agreement on how this revision may take place.” (Department of Health and Home Office, October 2008, op cit, chapter 9) And the Law Commission has recently weighed into the debate. As part of a much wider review of adult social care law, it says it wants to look for itself at the notion of vulnerability. (Law Commission, November 2008, Adult Social Care: Scoping Report, paragraphs 4.280-4.293)

Alternatives
So, what are the alternatives? The Law Commission speaks favourably of the Safeguarding Vulnerable Groups Act 2006, which, it says, understands vulnerability “purely through the situation an adult is placed [in]”. (Law Commission, 2008, op cit, paragraph 4.290) Some have argued, however, that it would be better to abandon the notion of vulnerability and instead, seek to protect people who are simply at risk.

That, certainly, is the purpose of legislation recently enacted in Scotland. The Adult Support and Protection (Scotland) Act 2007 covers people who “(1) are unable to safeguard their own well-being, property, rights or other interests; (2) are at risk of harm; and (3) because they are affected by disability, mental disorder, illness or physical or mental infirmity, are more vulnerable to being harmed than adults who are not so affected.”

ADASS, too, supports the use of risk as the keystone of adult protection, although its definition differs from the one used in Scotland. It says an adult at risk is one “who is or may be eligible for community care services” and whose independence and wellbeing are at risk due to abuse or neglect. (ADASS, 2005, op cit) This reference to community care need not fall foul of the Commons Health Committee, however, for ADASS says it includes “those people who are assessed as being able to purchase all or part of their community care services but whose need – in relation to safeguarding – is for access to mainstream services such as the police.” (Ibid)

It remains to be seen what the adult protection review will yield, and which notion the government will choose. But its task might be even more important than the consultation document suggests.

The stakes
With the coming of the Mental Capacity Act 2005 and the creation of a new Court of Protection, the High Court has lost the work it used to do with incapable people. Recently, however – and perhaps not coincidentally - it has set about transforming its inherent jurisdiction so as to offer protection to what it calls ‘vulnerable adults’. It proposes nothing less than the regulation of “everything that conduces to [their] welfare and happiness”. (Re SA (Vulnerable adult with capacity: Marriage) [2006] 1 FLR 867)

Conclusion
Adult protection is too important a task to be scuppered by questions of nomenclature. Everyone concerned needs to know precisely when the safeguards will kick in; when, it seems, an adult will be vulnerable. And the apparent willingness of the High Court to supplement the No Secrets guidance with real, enforceable rights means that as far as such people are concerned, the stakes have never been higher.

A real and immediate risk: clinical negligence and the right to life

Where health services and practitioners face claims under the Human Rights Act (HRA), they will usually have to show that they did all that could reasonably be expected of them.

That was the decision in a case involving a woman with paranoid schizophrenia, who was able to leave the hospital in which she was detained, walk two miles to a railway station and throw herself in front of a train. (Savage v South Essex Partnership NHS Foundation Trust [2008] UKHL 74)

The woman’s daughter wishes to bring proceedings against the NHS trust that manages the hospital. She says the trust breached her mother’s right to life under article 2 of the European Convention on Human Rights. She is unable to bring proceedings under either the Law Reform (Miscellaneous Provisions) Act 1934 or the Fatal Accidents Accident Act 1976, and therefore chose to pursue her claim via the HRA. Ultimately, she will have to establish that she is a ‘victim’ of the alleged breach, but for present purposes, that issue was not in play.

In a judgment delivered on 10 December 2008, the House of Lords said the daughter’s claim could proceed to trial, and it gave guidance as to the standard the trust will have to meet if it is to defend the proceedings successfully.

Their Lordships said the effect of article 2 is to impose two duties upon health bodies such as NHS trusts.

The first duty is to protect the lives of patients. Hospital trusts must ensure they employ competent staff, who are trained to a high professional standard, and adopt safe systems of work. If, for example, there was no proper system for supervising mentally ill patients - and if, as a result, a patient was able to commit suicide - the trust would have violated his right to life.

Even where there are competent staff and safe systems, of course, an individual doctor or nurse might act negligently and thereby cause a patient’s death. In that situation, the practitioner would be personally liable in damages. The health body might also be vicariously liable for the practitioner’s negligence, but it would not have breached article 2, because it had done all that the article requires of it to protect the patient’s life.

Their Lordships said the second duty imposed by article 2 is operational in nature and more specific than the first one. It arises only if members of staff know or ought to know that a particular patient presents a ‘real and immediate’ risk. In that circumstance, article 2 requires staff to do all that can reasonably be expected to do to protect the patient from the risk. If they fail in this regard, not only will they and the health body be liable in negligence; there will also be a violation of the obligation under article 2 to protect the patient’s life. It was accepted, however, that resources are not unlimited, and their Lordships said that in attempting to fulfil their obligation to a patient, practitioners need not behave any more cautiously than they were already persuaded to do by the ordinary law of negligence.

Traditionally, of course, the right to life has most often been engaged by deaths occurring in custody, with little or no distinction being made between prison and police custody and detention under the Mental Health Act. One judge, however, suggested that article 2 might extend further than that.

Baroness Hale – who, as well as being a senior judge, is also a noted mental health law expert – said she found it difficult to distinguish between compulsory hospital patients and those on formal leave, and also between detained and informal patients, and between them and the incapable patients who, from next April, will be subject to the Deprivation of Liberty Safeguards. All of them, she implied, might now be entitled to the protection of this special duty of care.

The primary function of this decision is to clarify the obligations imposed by the Human Rights Act: generally, it is unlikely to impose upon health care services and practitioners greater obligations than they already owed. But the decision does acknowledge a special duty in some circumstances, and in time, we may conclude that its greatest novelty was to extend those circumstances and, therefore, the number of patients protected by article 2.