United Kingdom Supreme Court

You are here: BAILII >> Databases >> United Kingdom Supreme Court >> Nicklinson & Anor R (on the application of) (Rev 1) [2014] UKSC 38 (25 June 2014)
URL: http://www.bailii.org/uk/cases/UKSC/2014/38.html
Cite as: [2014] HRLR 17, 36 BHRC 465, [2015] AC 657, [2015] 1 AC 657, 139 BMLR 1, [2014] 3 FCR 1, [2014] UKSC 38, [2014] WLR(D) 298, (2014) 139 BMLR 1, [2014] 3 WLR 200, [2014] 3 All ER 843

[New search] [Printable PDF version] [Buy ICLR report: [2015] AC 657] [Buy ICLR report: [2015] 1 AC 657] [Buy ICLR report: [2014] 3 WLR 200] [View ICLR summary: [2014] WLR(D) 298] [Help]

Summary

Trinity Term
[2014] UKSC 38
On appeal from: [2013] EWCA Civ 961

JUDGMENT

R (on the application of Nicklinson and another) (Appellants) v Ministry of Justice (Respondent)

R (on the application of AM) (AP) (Respondent) v The Director of Public Prosecutions (Appellant)

before

Lord Neuberger, President
Lady Hale, Deputy President
Lord Mance
Lord Kerr
Lord Clarke
Lord Wilson
Lord Sumption
Lord Reed
Lord Hughes

JUDGMENT GIVEN ON

25 June 2014

Heard on 16-19 December 2013

Appellant (Nicklinson and Lamb) Paul Bowen QC Guy Vassall-Adams (Instructed by Bindmans LLP)		Respondent (Ministry of Justice) David Perry QC James Strachan QC (Instructed by Treasury Solicitors)

Appellant (DPP) John McGuinness QC Duncan Atkinson (Instructed by CPS Appeals Unit)		Respondent (AM) Phillip Havers QC Adam Sandell (Instructed by Leigh Day & Co)

Intervener (CNK Alliance Ltd (Care Not Killing)) Charles Foster Ben Bradley Katarina Sydow (Instructed by Barlow Robbins LLP)		Intervener (Dignity and Choice in Dying) Lord Pannick QC Ravi Mehta (Instructed by Baker & McKenzie LLP)

Intervener (British Humanist Association) Heather Rogers QC Caoilfhionn Gallagher Maria Roche (Instructed by Irwin Mitchell LLP)

LORD NEUBERGER

These appeals arise out of tragic facts and raise difficult and significant issues, namely whether the present state of the law of England and Wales relating to assisting suicide infringes the European Convention on Human Rights, and whether the code published by the Director of Public Prosecutions ("the DPP") relating to prosecutions of those who are alleged to have assisted a suicide is lawful.

The appeals arise out of claims brought by three men, Tony Nicklinson, Paul Lamb and someone known for the purpose of these proceedings as Martin, each of whom was suffering such a distressing and undignified life that he had long wished to end it, but could not do so himself because of his acute physical incapacity. Mr Lamb contends that the law should permit him to seek assistance in killing himself in this country, and, if it does not, it should be changed so as to enable him to do so. He is supported by the widow of Mr Nicklinson, who has died since the proceedings were issued. Martin's case is that there should be clearer guidance in the policy published by the DPP with regard to prosecuting those from whom he would like advice and assistance in connection with killing himself.

An outline of the facts

The first appeal arises from the fact that Mr Nicklinson suffered a catastrophic stroke eight or nine years ago, when he was aged 51. As a result, he was completely paralysed, save that he could move his head and his eyes. He was able to communicate, but only laboriously, by blinking to spell out words, letter by letter, initially via a perspex board, and subsequently via an eye blink computer. Despite loving and devoted attention from his family and carers, his evidence was that he had for the past seven years consistently regarded his life as "dull, miserable, demeaning, undignified and intolerable", and had wished to end it.

Because of his paralysed state, Mr Nicklinson was unable to fulfil his wish of ending his life without assistance, other than by self-starvation, a potentially protracted exercise, involving considerable pain and distress. His preference was for someone to kill him by injecting him with a lethal drug, such as a barbiturate, but, if that was not acceptable, he was prepared to kill himself by means of a machine invented by Philip Nitschke, an Australian doctor. This machine, after being loaded with a lethal drug, could be set up so as to be digitally activated by Mr Nicklinson, using a pass phrase, via an eye blink computer.

Because he was told that it would be unlawful for someone to kill him or even to assist him in killing himself, Mr Nicklinson applied to the High Court for (i) a declaration that it would be lawful for a doctor to kill him or to assist him in terminating his life, or, if that was refused, (ii) a declaration that the current state of the law in that connection was incompatible with his rights under article 8 of the Convention.

While expressing great sympathy and respect for Mr Nicklinson's situation and wishes, the High Court, in an impressive judgment given by Toulson LJ, with whom Royce and Macur JJ agreed, refused him both forms of relief – [2012] EWHC 2381 (Admin). Following that decision, Mr Nicklinson embarked on the very difficult and painful course of self-starvation, refusing all nutrition, fluids, and medical treatment, and he died of pneumonia on 22 August 2012.

Mr Nicklinson's wife, Jane, was then both added (because she contended that she had a claim in her own right) and substituted (in her capacity as administratrix of Mr Nicklinson's estate) as a party to the proceedings, and pursued an appeal to the Court of Appeal. The Court of Appeal, while again sympathetic and respectful of her position, dismissed her appeal for reasons given in a similarly impressive judgment by Lord Dyson MR and Elias LJ, with whom Lord Judge CJ agreed – [2013] EWCA Civ 961; [2014] 2 All ER 32.

Because it was feared that there might be a challenge to Mrs Nicklinson's right to pursue an appeal, Paul Lamb was added as a claimant in the proceedings before the hearing in the Court of Appeal. Since a catastrophic car crash in 1990, Mr Lamb has been completely immobile, save that he is able to move his right hand. He requires carers 24 hours a day, suffers pain every day, and is permanently on morphine. His condition is irreversible, and he wishes a doctor to end his life, which he regards as consisting of a mixture of monotony, indignity and pain. He therefore applied for the same relief as Mr Nicklinson had sought, and it was similarly refused by the Court of Appeal.

The second appeal arises from the fact that Martin (who wishes to be so described in order to maintain his privacy) suffered a brainstem stroke in August 2008, when he was 43. He is almost completely unable to move and can only communicate thorough slow hand movements and via an eye blink computer. His condition is incurable, and, despite being devotedly looked after by his wife and carers, his evidence is that he wishes to end his life, which he regards as undignified, distressing and intolerable, as soon as possible.

Apart from self-starvation, Martin's only way of achieving this is by travelling to Zurich in Switzerland to make use of the Dignitas service, which, lawfully under Swiss law, enables people who wish to die to do so. However, he first needs (i) to find out about this service, (ii) to join Dignitas, (iii) to obtain his medical records, (iv) to send Dignitas money, and (v) to have someone accompany him to Zurich. For understandable reasons, his wife does not want to be involved, and he does not want to involve any other member of his family, in this project. So, as he says, he needs assistance from one of his carers or from an organisation such as Friends At The End.

Martin began proceedings seeking an order that the DPP should clarify, and modify, his published "Policy for Prosecutors in respect of Cases of Encouraging or Assisting Suicide", published in February 2010 ("the 2010 Policy") and other relief. He seeks the clarification and modification to enable responsible people, including, but not limited to, carers who are willing to do so, to know that they could assist Martin in committing suicide through Dignitas without the risk of being prosecuted.

Martin's proceedings were heard together with those brought by Mr Nicklinson, and they failed in the High Court. A few months later, he embarked on an attempt to end his life by self-starvation, but abandoned it in distressing circumstances. Martin's appeal, which was heard together with that of Mrs Nicklinson and Mr Lamb, was partially successful, in that Lord Dyson and Elias LJ considered that, "in certain respects", the 2010 Policy was "not sufficiently clear … in relation to healthcare professionals" – [2013] EWCA Civ 961, para 140; [2014] 2 All ER 32. Lord Judge CJ took a different view, and would have dismissed Martin's appeal.

The Court of Appeal gave Mrs Nicklinson and Mr Lamb ("the appellants") permission to appeal to the Supreme Court in the first appeal. In the second appeal, the Court of Appeal gave the DPP permission to appeal, and Martin permission to cross-appeal, as he contends that the order of the Court of Appeal in his case does not go far enough.

The tragic situations in which Mr Nicklinson, Mr Lamb and Martin found or find themselves are not as uncommon as some may like to think. There is reliable statistical and anecdotal evidence which indicates that, in recent years, hundreds of people suffering from terminal or chronic conditions, whose lives are often painful and/or undignified, committed suicide annually, that a significant number of them were assisted in doing so, and that there are many who wish to die, but (like Mr Nicklinson, Mr Lamb and Martin) cannot do so without assistance or advice, which it is generally assumed that they are unable to obtain because of the current state of the law. Examples of such evidence may be found in Assisted Dying for the Terminally Ill Bill – First Report HL Paper 86-I, 2005, especially para 77, and the Report on Assisted Dying, the Falconer Report, 2012, especially pp 108-138.

The legal and policy background

The domestic law relating to killing and suicide

Murder represents the most serious form of homicide, and it is a common law offence in England and Wales, although some of its ingredients have been altered by legislation, most significantly by the Homicide Act 1957 ("the 1957 Act"). For present purposes, it suffices to say that the offence of murder involves the perpetrator killing a person when intending either to kill or to inflict grievous bodily harm. A conviction for murder carries a mandatory life sentence.

Manslaughter is also a common law offence with statutory amendments, again most notably in the 1957 Act. The offence of voluntary (as opposed to involuntary) manslaughter is, in effect, murder in circumstances where the perpetrator is able to raise certain specified grounds of mitigation, including diminished responsibility and loss of control (all of which are subject to certain requirements). Manslaughter carries a maximum sentence of life imprisonment, and there is no minimum sentence.

Mercy killing is a term which means killing another person for motives which appear, at least to the perpetrator, to be well-intentioned, namely for the benefit of that person, very often at that person's request. Nonetheless, mercy killing involves the perpetrator intentionally killing another person, and therefore, even where that person wished to die, or the killing was purely out of compassion and love, the current state of the law is that the killing will amount to murder or (if one or more of the mitigating circumstances are present) manslaughter – see per Lord Judge CJ in R v Inglis [2011] 1 WLR 1110, para 37. As Lord Browne-Wilkinson said in Airedale NHS Trust v Bland [1993] AC 789, 885, "the doing of a positive act with the intention of ending life is and remains murder".

Nonetheless, a doctor commits no offence when treating a patient in a way which hastens death, if the purpose of the treatment is to relieve pain and suffering (the so-called "double effect") – see per Lord Goff of Chieveley in Bland at p 867. The House of Lords in that case decided that no offence was involved in refusing or withdrawing medical treatment or assistance, ultimately because this involved an omission rather than a positive act. While Lord Goff, Lord Browne-Wilkinson and Lord Mustill were all concerned about the artificiality of such a sharp legal distinction between acts and omissions in this context, they also saw the need for a line to be drawn, and the need for the law in this sensitive area to be clear – see at pp 865, 885 and 887 respectively.

Until 1961, it was an offence to commit suicide, which was regarded as self-murder; people who unsuccessfully attempted to kill themselves were not infrequently prosecuted. Section 1 of the Suicide Act 1961 ("the 1961 Act") provided that "[t]he rule of law whereby it is a crime for a person to commit suicide is hereby abrogated". As suicide was regarded as self-murder before 1961, a person who aided or encouraged another person to commit suicide committed an offence; thus, the survivor of a suicide pact was guilty of murdering the successful self-murderer – see R v Croft [1944] 1 KB 295. Section 4 of the 1957 Act provided that such a survivor would only be guilty of manslaughter. However, the abolition of suicide four years later as a crime meant that it was necessary to address the question of what to do about assisting and encouraging suicide.

Parliament dealt with that issue in section 2 of the 1961 Act ("section 2"), subsection (1) of which has now been repealed and re-enacted in the form of subsections (1)-(1C) by section 59(2) of the Coroners and Justice Act 2009 ("the 2009 Act"). The relevant parts of section 2 in its current form provide as follows:

"(1) A person ("D") commits an offence if—

(a) D does an act capable of encouraging or assisting the suicide or attempted suicide of another person, and

(b) D's act was intended to encourage or assist suicide or an attempt at suicide.

…

(1C) An offence under this section is triable on indictment and a person convicted of such an offence is liable to imprisonment for a term not exceeding 14 years.

….

(4) [N]o proceedings shall be instituted for an offence under this section except by or with the consent of the Director of Public Prosecutions."

The involvement of the civil courts

In Bland, the House of Lords held that it was lawful for doctors to discontinue treatment of a person who was in what was then called a persistent vegetative state. As Lord Goff explained at p 864, it had already been established that "if an adult patient of sound mind refuses, however unreasonably, to consent to treatment or care by which his life would or might be prolonged, the doctors responsible for his care must give effect to his wishes, even though they do not consider it to be in his best interests to do so". Where a person was unable to communicate his wishes, the correct question to ask, according to Lord Goff at p 868, was "whether it is in his best interests that treatment which has the effect of artificially prolonging his life should be continued", and in that case the answer was in the negative.

In adopting the "best interests" principle, the House of Lords followed its earlier decision in In re F (Mental Patient: Sterilisation) [1990] 2 AC 1, and in adopting the omission/commission distinction, it followed the approach of the Court of Appeal in two cases which raised the question of medical treatment for a severely disabled child - In re B (A Minor) (Wardship: Medical Treatment) [1981] 1 WLR 1421 ("Re B (Wardship)") and In re J (A Minor) (Wardship: Medical Treatment) [1991] Fam 33. Lord Goff accepted that there was a fundamental difference between a positive action which caused death and an omission which resulted in a death. At p 866, he said:

"[T]the doctor's conduct in discontinuing life support can properly be categorised as an omission. It is true that it may be difficult to describe what the doctor actually does as an omission, for example where he takes some positive step to bring the life support to an end. But discontinuation of life support is, for present purposes, no different from not initiating life support in the first place."

The way in which that passage is expressed indicates a certain and understandable discomfort with the notion that switching a machine off actually is an omission. A little later, Lord Goff dealt with another difficulty to which his conclusion gave rise, when he contrasted the position of a doctor in such a case with that of an "interloper who maliciously switches off a life support machine". Although he did not expressly say so, such an action must, I think, amount to murder or manslaughter, and Lord Goff dealt with the difficulty by saying that such an interloper would be "actively intervening to stop the doctor from prolonging the patient's life, and such conduct cannot possibly be categorised as an omission".

Subsequently, there has been a number of cases where, in the best interests of a patient, and often contrary to the wishes of his close family, the court has authorised switching off a life support machine, stopping providing food and drink, and withholding medical treatment (even of an elementary nature), all of which would lead inevitably to death. As was said in Bland, the common law has always recognised the right of a person to refuse treatment in advance, and, in that connection, Parliament has intervened to an extent through sections 24-26 of the Mental Capacity Act 2005, which permits individuals with capacity to make a valid advance direction refusing medical treatment, including treatment which would be life sustaining. Further, the courts have also recognised that, where a patient is unable to give her consent, it is lawful to give her treatment if it is necessary in her best interests – see Re F.

In cases of withdrawal of treatment, the House of Lords recommended in Bland that, before treatment could be withheld in any case where it was impossible for the patient to be consulted, permission should be sought from the High Court until "a body of experience and practice [had] buil[t] up which will obviate the need for application in every case" – pp 873-4. The role of the court in such cases was recently discussed by Lady Hale in Aintree University Hospitals NHS Foundation Trust v James [2013] 3 WLR 1299, paras 18-22 and 35-39.

As Hoffmann LJ said in Bland at p 825, "Modern medicine … faces us with fundamental and painful decisions about life and death which cannot be answered on the basis of normal everyday assumptions". The accuracy of this observation was subsequently demonstrated by the decision of the Court of Appeal In re A (Children) (Conjoined Twins: Surgical Separation) [2001] Fam 147 ("Conjoined Twins"). This decision took the law further in that the court authorised surgeons to separate conjoined twins, a positive act rather than omission, which would inevitably hasten the death of one twin in order to improve very considerably the life expectancy of the other.

In the subsequent case of In re B (Consent to Treatment – Capacity) [2002] 1 FLR 1090 ("Re B (Treatment)"), the applicant, who was effectively tetraplegic, and who was dependent on an artificial ventilation machine in order to breathe, wished the machine to be turned off, as she wanted to die, owing to the very poor quality of her life. Her doctors refused to turn the machine off, and she applied to the court for an order that they do so. Having concluded that the applicant had the mental capacity to make the decision, Dame Elizabeth Butler-Sloss P decided that the issue was not to be determined by considering what the court concluded was in her best interest. As explained in para 23 above, under the common law, it was purely a matter for the applicant whether or not the machine was turned off, provided that she was in a fit mental state to form a view. And, as she wanted the machine turned off, and she was mentally fit, the continued application of the machine to her body constituted in law trespass to the person. Accordingly she was granted the relief which she sought.

The Convention and assisted suicide

The two most central rights contained in the Convention for the purposes of the present appeals are in articles 2 and 8. Article 2, in summary form, guarantees the right to life, and, unsurprisingly, it is an unqualified right. Article 8.1 entitles everyone to "respect for his private … life". This right is qualified, as article 8.2 prohibits any "interference by a public authority with the exercise of this right" unless (i) "it is in accordance with the law", and (ii) it "is necessary in a democratic society, … for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others".

In R (Pretty) v Director of Public Prosecutions [2002] 1 AC 800 ("Pretty v DPP"), Mrs Pretty, who suffered from the progressive condition of motor neurone disease, complained that (i) the refusal of the DPP to grant her husband proleptic immunity from prosecution if he assisted her in killing herself (which she wished to do when her disease became intolerable), and/or (ii) the prohibition on assisting suicide in section 2, violated her rights under articles 2, 3, 8, 9 and 14 of the Convention. The House of Lords held that Mrs Pretty's desire to end her life prematurely did not engage her rights under any of those articles. The House went on to find that, if this was wrong, the government, to quote Lord Bingham at para 30, "ha[d] shown ample grounds to justify the existing law and the current application of it", although this was "not to say that no other law or application of it would be consistent with the Convention". This view was also adopted by Lord Steyn, Lord Hope, and Lord Scott at paras 62, 97, and 124, and, albeit implicitly, by Lord Hobhouse at paras 111 and 120.

Mrs Pretty then applied to the European Court of Human Rights ("the Strasbourg court"), where she was partially successful, in that it was held, albeit in somewhat guarded terms, that her desire to end her life did engage article 8.1, but not any other article - see Pretty v United Kingdom (2002) 35 EHRR 1 ("Pretty v UK"), para 67. In three subsequent decisions, the Strasbourg court has stated in clear terms that article 8.1 encompasses the right to decide how and when to die, and in particular the right to avoid a distressing and undignified end to life (provided that the decision is made freely) – see Haas v Switzerland (2011) 53 EHRR 33, para 51, Koch v Germany (2013) 56 EHRR 6, paras 46 and 51, and Gross v Switzerland (2014) 58 EHRR 7, para 60.

These cases also establish that the fact that a third party may have to be involved in enabling a person to die does not prevent that person from invoking article 8.1. Furthermore, it is clear from Koch, paras 43-46 that a person in Mrs Nicklinson's position, namely a spouse or partner who shares a close relationship with the person who wishes to die, and is closely involved in that person's suffering and desire to die, can invoke an article 8 right of her own in that connection. It is also clear from Koch, paras 78-82 that, at least in the Strasbourg court, Mrs Nicklinson would not be able to rely on her late husband's article 8 rights in her capacity as his personal representative or sole beneficiary.

Although Mrs Pretty's article 8 rights were held to have been interfered with in Pretty v UK, she failed in her claim, because the interference with her right was held to be justified by article 8.2, at least from the perspective of the Strasbourg court. In para 74 of its decision, the Strasbourg court described section 2 as "designed to safeguard life by protecting the weak and vulnerable and especially those who are not in a condition to take informed decisions against acts intended to end life or to assist in ending life". The court also said that many "terminally ill individuals … will be vulnerable, and it is the vulnerability of the class which provides the rationale for the law in question." Accordingly, it was "primarily for states to assess the risk and the likely incidence of abuse if the general prohibition on assisted suicides were relaxed or exceptions were to be created".

At para 76, the Strasbourg court said this:

"The Court does not consider therefore that the blanket nature of the ban on assisted suicide is disproportionate. The Government has stated that flexibility is provided for in individual cases by the fact that consent is needed from the DPP to bring a prosecution and by the fact that a maximum sentence is provided … . It does not appear arbitrary to the Court for the law to reflect the importance of the right to life, by prohibiting assisted suicide while providing for a system of enforcement and adjudication which allows due regard to be given in each particular case to the public interest in bringing a prosecution, as well as to the fair and proper requirements of retribution and deterrence."

The court accordingly concluded in para 78 that "the interference in this case may be justified as 'necessary in a democratic society' for the protection of the rights of others", so that there was no violation of article 8.

In Haas, the applicant was severely bipolar, and wanted to obtain a lethal dose of a drug to kill himself, but could not do so, because Swiss law required him to get a prescription, and, before he could do that, he needed a psychiatric assessment. The Strasbourg court referred at para 55 to the fact that "the vast majority of member states seem to attach more weight to the protection of the individual's life than to his or her right to terminate it", and therefore considered that "the states enjoy a considerable margin of appreciation in this area".

The court accordingly concluded in para 56, that, although it had sympathy with the applicant's wishes, "the regulations put in place by the Swiss authorities … pursue, inter alia, the legitimate aims of protecting everybody from hasty decisions and preventing abuse". The court also observed in para 58 that "the right to life guaranteed by article 2 … obliges states to establish a procedure capable of ensuring that a decision to end one's life does indeed correspond to the free wish of the individual concerned".

In Koch, the applicant's late wife, who was tetraplegic, needed his help to commit suicide. The Strasbourg court considered that the German courts' failure to entertain his application, which was for a declaration that the refusal of a Federal drugs institute to enable him to obtain a lethal dose of medication was unlawful, infringed his article 8 rights, which could "encompass a right to judicial review, even in a case in which the substantive right in question had yet to be established" – para 53. For present purposes, the case is of interest mainly because, in para 26, the court explained that in 36 of the 43 member states (including the UK) "any form of assistance to suicide is strictly prohibited and criminalised by law", in three (Germany, Sweden and Estonia) such "assistance is not a criminal offence", and four (Switzerland, Belgium, the Netherlands and Luxembourg) "allowed medical practitioners to prescribe lethal drugs, subject to specific safeguards". At para 70, the court stated that the fact that "the state parties to the Convention are far from reaching a consensus" on the legal treatment of assisting suicide "points to a considerable margin of appreciation enjoyed by the state in this context".

In Gross, the applicant had become so old and frail that she found her quality of life so bad that she had for some time wished to kill herself. However, she was unable to find a doctor in Switzerland who would provide her with the necessary prescription for a lethal drug, because her counsel was unable to guarantee that any doctor who prescribed the drug "would not risk any consequences from the point of view of the code of professional medical conduct" – para 11. At para 62, the court observed that there could be "positive obligations inherent in an effective 'respect' for private life", and that this could include "both the provision of a regulatory framework of adjudicatory and enforcement machinery protecting individuals' rights and the implementation, where appropriate, of specific measures". At para 63, the court explained that the applicant's case "primarily raises the question whether the State had failed to provide sufficient guidelines defining if and … under which circumstances medical practitioners were authorised to issue a medical prescription to a person in the applicant's circumstances".

Having considered the Swiss law on the topic, the court (in what was a bare majority judgment, as three of the seven judges dissented) held that the applicant's article 8 rights were infringed. The court said in para 65 that there was a "lack of clear legal guidelines", which was "likely to have a chilling effect on doctors who would otherwise be inclined to provide someone such as the applicant with the requested medical prescription". In the following paragraph, the court explained that, "if there had been clear, state-approved guidelines defining the circumstances under which medical practitioners are authorised to issue the requested prescription in cases where an individual has come to a serious decision, in the exercise of his or her free will, to end his or her life, but where death is not imminent as a result of a specific medical condition", the applicant would not have "found herself in a state of anguish and uncertainty regarding the extent of her right to end her life".

So far as the domestic position is concerned, section 1 of the Human Rights Act ("the 1998 Act") defines "Convention rights" as, inter alia, the rights set out in articles 2-12 and 14 of the Convention. Section 3(1) provides that "[s]o far as it is possible to do so, … legislation must be read and given effect in a way which is compatible with the Convention rights". Section 4 states that where one of the more senior courts in the UK concludes that a statutory provisions is nonetheless "incompatible with a Convention right, … it may make a declaration of that incompatibility". Section 6 requires public authorities to act compatibly with the Convention save where statute prevents them from doing so.

The role of the DPP

Section 2(4) of the 1961 Act precludes any prosecution of a person who has allegedly contravened section 2(1) without the DPP's consent. However, as Lord Hughes convincingly demonstrates in his judgment, section 2(4) has a relatively limited function. The DPP always has the right to decide that it is not in the public interest to prosecute, even where it is clear that an offence was committed; and the DPP has power to stay a private prosecution if satisfied, inter alia, that it is not in the public interest for the prosecution to proceed. All that section 2(4) does, therefore, is to rule out the bringing of a private prosecution for encouraging or assisting a suicide without the DPP's prior consent (although it is worth noting that, before the creation of the Crown Prosecution Service ("CPS"), it would have prevented the police prosecuting without the consent of the DPP). However, that does not undermine the importance of the prosecutorial discretion in connection with assisting suicide. The public importance of, and the public concern about, this discretion in the present context were recognised by the DPP in December 2008, when he voluntarily published a decision containing his full reasons for not prosecuting the parents of a tetraplegic young man for taking their son to Zurich to enable him to be assisted to kill himself, as discussed by Lord Hope and Lord Brown in R (Purdy) v Director of Public Prosecutions [2010] 1 AC 345, paras 49-51 and 79-81 respectively.

The proceedings in Purdy were brought following the decision of the Strasbourg court in Pretty v UK, in order to require the DPP to spell out his policy in relation to his prosecutorial discretion in a public document. Ms Purdy suffered from progressive multiple sclerosis and expected that a time would come when she would regard her continued existence as intolerable and would wish to end her life. She would need the assistance of her husband to do so (by taking her to Switzerland to enable her to use the services of Dignitas) and wished to ensure, as far as possible, that he would not be prosecuted under section 2(1) of the 1961 Act.

She sought information from the DPP as to his likely attitude to a prosecution of her husband in those circumstances, and he declined to give it. While maintaining her claim for information, Ms Purdy accepted that the DPP could not give her husband "a guarantee of immunity from prosecution", as this "would be a matter for Parliament" (per Lord Hope at para 30). Departing from its decision in Pretty v DPP, following the Strasbourg court's decision in Pretty v UK, the House of Lords upheld her contention that the DPP's refusal infringed her article 8 rights. Given that her article 8 rights were engaged, Ms Purdy was entitled to expect the law to be accessible and foreseeable, and this required that "the law must indicate with sufficient clarity the scope of any such discretion conferred on the competent authorities and the manner of its exercise", as Lord Hope said at para 43 quoting from Hasan and Chaush v Bulgaria (2002) 34 EHRR 55, para 84. The Strasbourg court also observed that "[t]he level of precision required of domestic legislation … depends to a considerable degree on the content of the instrument in question, the field it is designed to cover and the number and status of those to whom it is addressed".

The DPP's argument in Purdy was that his Code for Crown Prosecutors, issued under section 10 of the Prosecution of Offences Act 1985, provided sufficient guidance, but the House rejected this argument as the Code applied to all crimes and "[fell] short of what [was] needed to satisfy the Convention tests of accessibility and foreseeability" in relation to assisting a suicide – per Lord Hope at para 53. As Lady Hale put it in para 64, "the object of the exercise should be to focus, not upon a generalised concept of 'the public interest', but upon the features which will distinguish those cases in which deterrence will be disproportionate from those cases in which it will not". Accordingly, as Lord Hope said at para 56, the DPP should be required to "promulgate an offence-specific policy identifying the facts and circumstances which he will take into account in deciding, in a case such as that which Ms Purdy's case exemplifies, whether or not to consent to a prosecution".

Within three months of this decision, the DPP issued a draft policy, identifying sixteen factors which would favour prosecution, and thirteen which would point against prosecution. Eight of the sixteen and seven of the thirteen were said to carry more weight than the remaining eight and six respectively. The CPS consulted widely about the contents of this draft policy, raising a large number of questions, and receiving over 4700 responses, which the DPP describes as being of "a high quality" and "the largest number of responses the CPS has ever received about a single topic". As a result, he modified the draft policy and produced the 2010 Policy.

The 2010 Policy

The 2010 Policy is detailed. After making a number of points, including the need for a prosecutor to be satisfied that a case satisfies the evidential requirement before considering whether it satisfies the public interest requirement, it deals with the relevant public interest factors from para 39.

Para 39 makes the points that each case must be determined on its own merits, and that an overall assessment is required, a point repeated at para 47, where it is also stated that the list of factors in the 2010 Policy is not intended to be exhaustive. Para 39 also states that sometimes a single factor one way will outweigh a number of factors the other way, and para 40 points out that the absence of a specified factor should be regarded as neutral. Paras 41 and 42 deal with the reliability of the evidence relating to the factors.

The 2010 Policy then turns to "Public interest factors tending in favour of prosecution" and continues:

"43. A prosecution is more likely to be required if:

1. The victim was under 18 years of age;

2. The victim did not have the capacity (as defined by the Mental Capacity Act 2005) to reach an informed decision to commit suicide;

3. The victim had not reached a voluntary, clear, settled and informed decision to commit suicide;

4. The victim had not clearly and unequivocally communicated his or her decision to commit suicide to the suspect;

5. The victim did not seek the encouragement or assistance of the suspect personally or on his or her own initiative;

6. The suspect was not wholly motivated by compassion; for example, the suspect was motivated by the prospect that he or she or a person closely connected to him or her stood to gain in some way from the death of the victim;

7. The suspect pressured the victim to commit suicide;

8. The suspect did not take reasonable steps to ensure that any other person had not pressured the victim to commit suicide;

9. The suspect had a history of violence or abuse against the victim;

10. The victim was physically able to undertake the act that constituted the assistance him or herself;

11. The suspect was unknown to the victim and encouraged or assisted the victim to commit or attempt to commit suicide by providing specific information via, for example, a website or publication;

12. The suspect gave encouragement or assistance to more than one victim who were not known to each other;

13. The suspect was paid by the victim or those close to the victim for his or her encouragement or assistance;

14. The suspect was acting in his or her capacity as a medical doctor, nurse, other healthcare professional, a professional carer [whether for payment or not], or as a person in authority, such as a prison officer, and the victim was in his or her care;

15. The suspect was aware that the victim intended to commit suicide in a public place where it was reasonable to think that members of the public may be present;

16. The suspect was acting in his or her capacity as a person involved in the management or as an employee (whether for payment or not) of an organisation or group, a purpose of which is to provide a physical environment (whether for payment or not) in which to allow another to commit suicide.

44. On the question of whether a person stood to gain, (paragraph 43(6) see above), the police and the reviewing prosecutor should adopt a common sense approach. It is possible that the suspect may gain some benefit - financial or otherwise - from the resultant suicide of the victim after his or her act of encouragement or assistance. The critical element is the motive behind the suspect's act. If it is shown that compassion was the only driving force behind his or her actions, the fact that the suspect may have gained some benefit will not usually be treated as a factor tending in favour of prosecution. However, each case must be considered on its own merits and on its own facts."

The 2010 Policy then turns to "Public interest factors tending against prosecution", and continues:

"45. A prosecution is less likely to be required if:

1. The victim had reached a voluntary, clear, settled and informed decision to commit suicide;

2. The suspect was wholly motivated by compassion;

3. The actions of the suspect, although sufficient to come within the definition of the offence, were of only minor encouragement or assistance;

4. The suspect had sought to dissuade the victim from taking the course of action which resulted in his or her suicide;

5. The actions of the suspect may be characterised as reluctant encouragement or assistance in the face of a determined wish on the part of the victim to commit suicide;

6. The suspect reported the victim's suicide to the police and fully assisted them in their enquiries into the circumstances of the suicide or the attempt and his or her part in providing encouragement or assistance."

The DPP's evidence in these proceedings is that there has been only one prosecution under section 2, and that was a successful prosecution of someone who provided petrol and a lighter to a vulnerable man known to have suicidal intent, and who subsequently suffered severe burns as a result. The DPP also informed the Court that it appears from Dignitas's website that, between 1998 and 2011, a total of 215 people from the UK used its services, and that nobody providing assistance in that connection has been prosecuted.

Assisted dying: the debate

In Pretty v DPP at para 54, Lord Steyn explained that "the subject of euthanasia and assisted dying have been deeply controversial" for a very long time, and continued:

"The arguments and counter arguments have ranged widely. There is a conviction that human life is sacred and that the corollary is that euthanasia and assisted suicide are always wrong. This view is supported by the Roman Catholic Church, Islam and other religions. There is also a secular view, shared sometimes by atheists and agnostics, that human life is sacred. On the other side, there are many millions who do not hold these beliefs. For many the personal autonomy of individuals is predominant. They would argue that it is the moral right of individuals to have a say over the time and manner of their death. On the other hand, there are utilitarian arguments to the contrary effect. The terminally ill and those suffering great pain from incurable illnesses are often vulnerable. And not all families, whose interests are at stake, are wholly unselfish and loving. There is a risk that assisted suicide may be abused in the sense that such people may be persuaded that they want to die or that they ought to want to die. Another strand is that, when one knows the genuine wish of a terminally ill patient to die, they should not be forced against their will to endure a life they no longer wish to endure. Such views are countered by those who say it is a slippery slope or the thin end of the wedge. It is also argued that euthanasia and assisted suicide, under medical supervision, will undermine the trust between doctors and patients. It is said that protective safeguards are unworkable. The countervailing contentions of moral philosophers, medical experts and ordinary people are endless. The literature is vast …. It is not for us, in this case, to express a view on these arguments. But it is of great importance to note that these are ancient questions on which millions in the past have taken diametrically opposite views and still do."

Following the decision in Bland, the House of Lords Committee on Medical Ethics, after receiving evidence, reported that "[a]s far as assisted suicide is concerned", they saw "no reason to recommend any change in the law" (see HL Paper 21-I, 1994, para 26). This was primarily based on "the message which society sends to vulnerable and disadvantaged people", which "should not, however obliquely, encourage them to seek death, but should assure them of our care and support in life" (ibid, para 239). The Government in its response agreed on the grounds that a change in the law "would be open to abuse and put the lives of the weak and vulnerable at risk" – (1994) Cm 2553, page 5.

The possibility of relaxing the statutory prohibition on assisting suicide has been debated in the House of Lords and House of Commons on at least six occasions in the past nine years. Thus, in November 2005, following the publication of HL Paper 86-1 referred to in para 14 above, Lord Joffe unsuccessfully introduced the Assisted Dying for the Terminally Ill Bill ("the 2005 Bill") in the House of Lords, and in July 2009, Lord Falconer of Thoroton moved an amendment that would have permitted assisting the "terminally ill" to commit suicide during the debate on the Bill which became the Coroners and Justice Act 2009. During the debate on the 2005 Bill, Lord Joffe made it clear that he did not "support assisted dying for patients who are not terminally ill", and that this was reflected in the Bill, on the basis that "after three years of legislative effort on the subject, I have no intention of pursuing this issue beyond the ambit of the present Bill" – Hansard (HL Debates), 12 May 2006 Col 1188. During the July 2009 debate on the Bill which became the 2009 Act, the House of Lords defeated the amendment – Hansard (HL Debates) 7 July 2009, cols 595ff. Their Lordships approved section 59 of the 2009 Act, whose purpose, as explained above, was to re-enact section 2 of the 1961 Act in clearer terms.

There was an adjournment debate on assisted dying in the House of Commons in November 2008 - Hansard (HC Debates), 11 November 2008, cols 221WHff. The House of Commons also approved the 2009 Act in a brief debate during which the purpose of section 59 was explained – Hansard (HC Debates), 26 January 2009, col 35. More recently, there was a debate on the Director's 2010 Policy in the House of Commons in March 2012, where changes in the law were mooted, but the 2010 Policy was approved – see Hansard (HC Debates), 27 March 2012, cols 1363ff.

In September 2010, Lord Falconer set up and chaired a commission on "Assisted Dying", which took evidence from many individuals and organisations, and the commission's report was published in January 2012. While it is a full and apparently balanced report, Lord Falconer is a strong and public supporter of liberalising the law on assisted dying, much of the funding of the commission came from people who take the same view, and some people who were against assisted dying refused to give evidence to the commission. The evidence from doctors and other caring professionals was mixed. The views of the medical professional bodies was also mixed - ranging from being against doctor involvement, via neutral to being in favour of it.

The Falconer Report indicated that in three jurisdictions where it was permissible to assist suicide, there was no evidence of vulnerable groups being subject to any pressure or coercion to seek an assisted death. The same view was expressed in the 2011 report of the Royal Society of Canada Expert Panel on End-of-Life Decision Making and in the 2012 report of the Quebec National Assembly 'Dying with Dignity' Select Committee. The Falconer Report concluded that "there [was] a strong case for providing the choice of assisted dying for terminally ill people", while "protecting" them and vulnerable people generally "from the risk of abuse or indirect social pressure to end their lives". However, the members of the Commission were "unable to reach a consensus on the issue of whether a person who has suffered a catastrophically life-changing event that has caused them to be profoundly incapacitated should be able to request an assisted death", but they were agreed that people who assisted loved ones and friends in that situation "should continue to be treated by the law with compassion and understanding".

The issues in these appeals

Introductory

In the first appeal, the appellants, Mrs Nicklinson and Mr Lamb, contend that section 2(1) of the 1961 Act, at least if read in accordance with conventional principles, constitutes a disproportionate, and therefore an unjustifiable, interference with the article 8 rights of people who have made "a voluntary, clear, settled and informed decision to commit suicide", and, who, solely because of their physical circumstances, require the assistance of a third party to achieve that end. I will refer to such people as "Applicants", a neutral and convenient, if not entirely accurate, expression.

The appellants' case is that the article 8 rights of Applicants should be accommodated by their being able to seek the assistance of third parties to enable them to kill themselves in a dignified and private manner, at a time of their choosing, in the United Kingdom, subject to some appropriate form of control so as to ensure that their decision to commit suicide is indeed voluntary, clear, settled and informed. Accordingly, they bring these proceedings against the Secretary of State for Justice, contending that this Court should either (i) read section 2(1) in such a way as to enable it to comply with the Convention (under section 3 of the 1998 Act), or, if that is not possible, (ii) make a declaration that section 2 is incompatible with the Convention (under section 4 of the 1998 Act). The Secretary of State contends that, in the light of the Strasbourg jurisprudence, this is not a contention which is capable of properly being raised before a court, and, even if that is wrong, bearing in mind the practical, moral and policy issues involved, this is not a contention which a domestic court should entertain under the United Kingdom's constitutional settlement.

Martin's argument in the DPP's appeal in the second appeal is rather different in its target. Although he also relies on article 8, Martin does not challenge the compatibility of section 2 with the Convention. His first argument is that the terms of the 2010 Policy are insufficiently clear in relation to the likelihood of prosecution of those individuals (other than relatives and close friends of the person concerned), especially including doctors and other members of the caring professions, who might otherwise be prepared, out of compassion, to provide a person who has a voluntary, clear, settled and informed wish to commit suicide, with information, advice and assistance in connection with that wish. His second argument is that the Policy should be modified to make it clear that, at any rate absent any aggravating circumstances, such an individual would not be liable to be prosecuted. The DPP argues that it would be inappropriate for a court to seek to dictate what her policy should be.

The first appeal raises the following issues:

a. Does section 2 impose an impermissible "blanket ban" on assisted suicide, outside the UK's permitted margin of appreciation? If not,

b. Given that the Strasbourg court has decided that it is for the member states to decide whether their own law on assisted suicide infringes article 8, does this Court have the constitutional competence to decide whether section 2 infringes article 8? If so,

c. Bearing in mind the nature of the issue, is it nonetheless inappropriate for this Court to consider whether section 2 infringes article 8, on the ground that it is an issue which is purely one for Parliament? If not,

d. In the light of the evidence and the arguments presented on this appeal, should the Court decide that section 2 infringes article 8? And finally,

e. In the light of the answers to these questions, what is the proper order to make on the first appeal?

It is perhaps worth explaining at this stage the difference between issues (b) and (c). Issue (b) raises the general question whether, in a case where the Strasbourg court decides that a point is within a member state's margin of appreciation, it is open to a domestic court to declare that a statutory provision, which is within that margin, nonetheless infringes Convention rights in the United Kingdom. Issue (c), which only arises if the court does have such power, is more specific to this case; it is whether, bearing in mind the nature of the point raised in the first appeal, a domestic court is an appropriate forum for considering whether the statutory provision involved, section 2 of the 1961 Act, infringes Convention rights in the United Kingdom, or whether the issue is best left entirely to Parliament. The second issue may be said to raise a constitutional point, whereas the third issue involves more of an institutional point.

The second appeal raises two points, namely:

f. (raised by the DPP's appeal) does the 2010 Policy comply with the requirements of article 8, and hence section 6 of the 1998 Act, and in particular the requirement of foreseeability? and

g. (raised by Martin's cross-appeal) if the DPP were to prosecute in a case such as Martin's, would it represent a disproportionate interference with his article 8 rights?

I shall deal with these issues, some of which have more than one facet, in turn.

Is section 2 within the UK's margin of appreciation under article 8? – issue (a)

The appellants contend, as a self-contained point, that the effect of the four Strasbourg court decisions on assisted suicide is that a "blanket ban" such as that imposed by section 2 infringes article 8, even allowing for the wide margin of appreciation accorded to member states. In other words, the appellants argue that, even allowing for the wide margin of appreciation afforded to member states on the issue of assisted suicide, a blanket ban would be regarded by the Strasbourg court as impermissibly outside that wide margin. This contention is said to be supported by the more general proposition that, where a ban curtails a Convention right, the Strasbourg court would hold that it cannot be a blanket ban. In support of this proposition, the appellants cite Hirst v UK (2006) 42 EHRR 41 which was concerned with the right of prisoners to vote.

I do not accept this argument. So far as the general point is concerned, the expression "blanket ban" is not helpful, as everything depends on how one defines the width of the blanket. Thus, a blanket ban on voting for all those serving life sentences would appear to be acceptable to the Strasbourg court – and certainly should be in my view. As for the more specific point, I do not consider that the Strasbourg jurisprudence suggests that a blanket ban on assisted dying is outside the margin of appreciation afforded to member states and, even if it is, then, in any event, the provisions of section 2(4) prevent the ban in this jurisdiction being a "blanket ban".

In connection with the specific point, the opening two sentences of para 76 of the Strasbourg court's decision in Pretty v UK (quoted in para 32 above) are not particularly happily worded. However, it appears to me that the effect of that decision is that, so far as the Strasbourg court is concerned, a national blanket ban on assisted suicide will not be held to be incompatible with article 8. The word "therefore" in the first sentence refers back to what precedes the paragraph, which (ignoring the discursion in para 75) is a passage at the end of para 74, which seems to me to say that it is a matter for each member state whether, and if so in what form, to provide for exceptions to a "general prohibition on assisted suicides". This conclusion is, I think, strongly supported by the fact that the court stated that the great majority of member states have what the appellants would characterise as blanket bans on assisted suicide.

The decision in Koch is said by the appellants to support the notion that a blanket ban on assisting a suicide cannot comply with article 8. I do not accept that. The question whether the German substantive law relating to the provision of prescriptions infringed article 8 was specifically left open, and the decision was limited to the fact that the applicant's article 8 rights had been infringed by the German court's refusal to consider that issue – see paras 52 and 71 of the judgment. Further, the Strasbourg court also made it clear in paras 70-71 that it was for the national court to decide whether what was effectively a prohibition on prescribing drugs to enable people to kill themselves infringed article 8, which appears to me to indicate that such a prohibition did not give rise to a problem under article 8 so far as the Strasbourg court was concerned.

Accordingly, I would reject the argument that a "blanket ban" on assisting suicide is outside the margin of appreciation afforded by the Strasbourg court to member states. In any event, it seems to me that, even if this is wrong, there can be no question of the Strasbourg court holding that section 2 infringes article 8 on the ground that it contains a blanket ban. What it said in paras 76-78 of Pretty v UK appears to me to make it clear that, whatever argument might have been raised if section 2(1) had stood on its own, prosecutorial discretion - reinforced by section 2(4), provided that it was implemented so as to render the law accessible and foreseeable, ensured that the current UK law relating to assisted suicide complied with the Convention so far as the Strasbourg court was concerned. None of the subsequent three decisions of that court on assisted suicide call this conclusion into question. (Of course, this would not mean that every aspect of the implementation of national law on assisted dying would be outside the scope of the Strasbourg court's consideration – cf the decisions in Koch and Gross).

Is it constitutionally open to the UK courts to consider compatibility? – issue (b)

The Strasbourg court explained in Pretty v UK, para 74, and Haas, para 57, that, when considering legislation on assisted suicide, one has to balance the article 8.1 rights of those who wish to be so assisted, against the need to protect the weak and vulnerable in relation to their article 2 and article 8.1 rights. The court has also acknowledged that views as to where the balance should come down can vary (eg in Gross, para 66), and that this is reflected by the different approaches in different members states - see Haas, para 55 and Koch, paras 26 and 70. As explained, this has led the Strasbourg court to conclude that member states enjoy a wide margin of appreciation on the issue of assisted dying – see Pretty v UK, para 74, Haas, para 55 and Koch, paras 70 and 71.

At first sight, it may appear from this that, as the High Court held, it would be inappropriate for this Court even to consider whether it should determine whether or not section 2 is incompatible with article 8. In R (Ullah) v Special Adjudicator [2004] 2 AC 323, para 20, Lord Bingham said that "[t]he duty of national courts is to keep pace with the Strasbourg jurisprudence as it evolves over time: no more, but certainly no less", and the Strasbourg court has determined that, at any rate so far as its jurisdiction is concerned, section 2 is consistent with the Convention. Accordingly, it might seem to follow that a UK court should not take a different view. It was, in part, on this basis that the Court of Appeal rejected the contention that section 2 was inconsistent with article 8 – see at [2013] EWCA Civ 961, paras 111-114; [2014] 2 All ER 32.

In my judgment, however, that is not a good answer to the claims made by the appellants. Lord Bingham's observation in Ullah was directed to the majority of cases raising claims that Convention rights have been infringed, where the Strasbourg court concludes either that there has been an infringement or that there has been no infringement. In such cases, in so far as they are capable of being of wider application than to the particular case before it, the Strasbourg court would intend that its conclusions and reasoning be applicable to all member states.

So far as the law on assisted suicide is concerned, the conclusion reached by the Strasbourg court is of a different nature. As explained above, the court has held that there is a wide margin of appreciation accorded to each state in this area, and that it is for each state to decide for itself how to accommodate the article 8 rights of those who wish and need to be assisted to kill themselves with the competing interests of the prevention of crime and the protection of others – see Pretty v UK, para 74, Haas, para 55 and Koch, paras 70 and 71. In those circumstances, it does not appear to me that the dictum quoted above from Ullah is in point. (For this reason, this is not the occasion to address the question whether, and if so how far, the principle enunciated by Lord Bingham in Ullah, para 20, should be modified or reconsidered.) In a case such as this, the national courts therefore must decide the issue for themselves, with relatively unconstraining guidance from the Strasbourg court, albeit bearing in mind the constitutional proprieties and such guidance from the Strasbourg jurisprudence, and indeed our own jurisprudence, as seems appropriate.

Support for this conclusion is to be found in In re G (Adoption: Unmarried Couple) [2009] 1 AC 173. In paras 33-35, Lord Hoffmann pointed out that "Convention rights", as defined in section 1 of the 1998 Act, were "domestic and not international rights", and that the duty of domestic courts under section 2 of that Act was to "take into account", rather than to regard themselves as bound by, decisions of the Strasbourg court, but that there were normally "good reasons why we should follow the interpretation adopted in Strasbourg".

At para 36 of re G, however, Lord Hoffmann said that different "considerations … apply in cases in which Strasbourg has deliberately declined to lay down an interpretation for all member states, as it does when it says the question is within the margin of appreciation". In the following paragraph, Lord Hoffmann stated that in such cases, "it is for the court in the United Kingdom to interpret [the relevant article or articles of the Convention] and to apply the division between the decision-making powers of courts and Parliament in the way which appears appropriate for the United Kingdom". He expanded on this by adding that "[t]he margin of appreciation is there for division between the three branches of government according to our principles of the separation of powers".

Lord Hope agreed with Lord Hoffmann at para 50, and Lady Hale expressed similar views at paras 116-120, saying pithily that "if the matter is within the margin of appreciation which Strasbourg would allow us, then we have to form our own judgment". Lord Mance, at para 130, took the same view, explaining that when "performing their duties under sections 3 and 6 [of the 1998 Act], courts must of course give appropriate weight to considerations of relative institutional competence". Having then emphasised the importance of giving weight "to the decisions of a representative legislature and democratic government within the discretionary area of judgment accorded to those bodies", he made the point that "the precise weight will depend, inter alia, on the nature of the right" and the extent to which it "falls within an area in which the legislature, executive, or judiciary can claim particular expertise". As Lord Hoffmann and Lord Mance explained, their approach does not involve the court calling into question the sovereignty of Parliament. The court has jurisdiction to consider whether a provision such as section 2 is compatible, or can be rendered compatible, with article 8, because that is part of the court's function as determined by Parliament in the 1998 Act. As it happens, it also reflects what the Strasbourg court decided about an individual's right of access to the court in Koch.

In an interesting passage in para 229 below, Lord Sumption suggests that, where an issue has been held by the Strasbourg court to be within the margin of appreciation, the extent to which it is appropriate for a UK court to consider for itself whether the Convention is infringed by the domestic law may depend on the reason why the Strasbourg court has concluded that the issue is within the margin. I agree that the reasoning of the Strasbourg court must be taken into account and accorded respect by a national court when considering whether the national law infringes the Convention domestically, in a case which is within the margin of appreciation – just as in any other case as section 2(1)(a) of the 1998 Act recognises. However, both the terms of the 1998 Act (in particular sections 2(1) and 4) and the principle of subsidiarity (as expounded for instance in Greens and MT v UK [2010] ECHR 1826, para 113) require UK judges ultimately to form their own view as to whether or not there is an infringement of Convention right for domestic purposes.

It is true that in Re G, the House of Lords was concerned with a statutory instrument, but the passages to which I have referred must, as a matter of logic and principle, be as applicable to primary, as to secondary, legislation. It is also true that the decision in Re G was based on the irrationality of the legislation concerned. Where the legislature has enacted a statutory provision which is within the margin of appreciation accorded to members states, it would be wrong in principle and contrary to the approach adopted in Re G, for a national court to frank the provision as a matter of course simply because it is rational. However, where the provision enacted by Parliament is both rational and within the margin of appreciation accorded by the Strasbourg court, a court in the United Kingdom would normally be very cautious before deciding that it infringes a Convention right. As Lord Mance said in Re G, the extent to which a UK court should be prepared to entertain holding that such legislation is incompatible must depend on all the circumstances, including the nature of the subject-matter, and the extent to which the legislature or judiciary could claim particular expertise or competence.

In these circumstances, given that the Strasbourg court has held that it is for each state to consider how to reconcile, or to balance, the article 8.1 rights of a person who wants assistance in dying with "the protection of … morals" and "the protection of the rights and freedoms of others", I conclude that, even under our constitutional settlement, which acknowledges parliamentary supremacy and has no written constitution, it is, in principle, open to a domestic court to consider whether section 2 infringes article 8. The more difficult question, to which I now turn, is whether we should do so.

Is it institutionally appropriate to consider whether section 2 infringes article 8? – issue (c)

Introductory

Having concluded that the court does have jurisdiction in principle to determine whether section 2 infringes the Convention, the next question is whether it is institutionally appropriate for a domestic court to consider whether section 2 infringes the article 8 rights of individuals such as Mr Nicklinson and Mr Lamb. In that connection, I have summarised the nub of their case in para 55 above.

In approaching this question, it is important to bear in mind that, as Lord Mance explained in Re G, what we have to consider is the breadth of the discretion which the courts should accord to Parliament, or, to put it another way, the limits of the courts' deference to Parliament's judgment, on the issue of the extent to which assisting suicide should be criminalised.

A summary of the parties' respective contentions

Section 2 interferes with the article 8 right of Applicants (as I have called them) to determine how and when they should die. Accordingly, it can only be a valid interference if it satisfies the requirements of article 8.2, ie if it is "necessary in a democratic society" for one or more of the purposes specified in that article, which in the present context would be "for the prevention of disorder or crime, for the protection of health or morals, or", most importantly for present purposes, "for the protection of the rights and freedoms of others".

When considering whether legislative measures satisfy those requirements, "four questions generally arise", as Lord Wilson explained in R (Aguilar Quila) v The Secretary of State for the Home Department [2012] 1 AC 621, para 45 (as recently illuminatingly discussed by Lord Reed in Bank Mellat v HM Treasury (No 2) [2013] 3 WLR 179, 222, paras 20ff):

"(a) is the legislative objective sufficiently important to justify limiting a fundamental right?; (b) are the measures which have been designed to meet it rationally connected to it?; (c) are they no more than are necessary to accomplish it?; and (d) do they strike a fair balance between the rights of the individual and the interests of the community?"

The appellants accept that "the legislative objective" of section 2 is to safeguard life, and in particular the lives of the vulnerable and the weak, including those "who are not in a position to take informed decisions against acts intended to end life or assist in ending life", to quote from Pretty v UK, para 74, or, as Lady Hale put it in Purdy at para 65, "people who are vulnerable to all sorts of pressures, both subtle and not so subtle, to consider their own lives a worthless burden to others".

As to the four requirements, as I will call them, identified in Lord Wilson's analysis, the appellants accept that requirement (a) is satisfied in that this objective is "sufficiently important to justify limiting a fundamental right", namely the article 8 right of those wish to end their lives and need the assistance of others to do so. They also accept that, so far as requirement (b) is concerned, section 2 has "been designed to meet" this objective and is "rationally connected to it". Accordingly, the issue whether section 2 infringes article 8 turns on whether requirements (c) and (d), necessity and balance, are satisfied.

In that connection, the appellants' case is that the absolute terms of section 2 are "more than necessary" to achieve its end, or that they do not strike a fair balance between the interests of Applicants and those of the weak and vulnerable, bearing in mind the grave and significant interference which it involves with the article 8 rights of Applicants, and that this is an argument which a domestic court should consider.

In summary terms, the Secretary of State's case is that, given that it is accepted that the statutory ban on assisting suicide, subject to prosecutorial discretion, can be rationally justified by the need to protect the weak and vulnerable and was recently affirmed by Parliament in the 2009 Act, any question of decriminalisation should be left to Parliament, as it is a controversial, difficult and sensitive moral and politico-social issue, which requires the assessment of many types of risk and the imposition of potentially complex regulations, and it is not a matter on which judges are particularly well informed or experienced. The Secretary of State also relies on the fact that section 2 was held to comply with the Convention by the House of Lords in Pretty v DPP less than thirteen years ago.

The protection of the weak and vulnerable

Although, as mentioned above, the appellants accept the Secretary of State's contention that section 2 is "designed to meet" the objective of protecting the weak and vulnerable and is "rationally connected to that objective", it is worth examining that contention. So far as assisting (as opposed to encouraging) suicide is concerned, section 2 is a somewhat indirect and blunt instrument in that it is, as a matter of practice, aimed at those who need assistance in committing suicide rather than those who are weak and vulnerable. It is a measure of the relative weakness of the connection that, in para 350 below, Lord Kerr concludes that, contrary to the appellants' concession, requirement (b) is not satisfied. I do not agree with that conclusion, because it seems to me, in general terms, that a blanket ban on assisting suicide will protect the weak and vulnerable, and, more particularly, that it may well be that those who are in the same unhappy position as Applicants, but do not wish to die, are in a particularly vulnerable position. However, the somewhat tenuous connection between the actual and intended targets is not irrelevant when one turns to requirements (c) and (d).

More specifically, if one concentrates on the appellants' argument that section 2 should be modified so as to exclude Applicants, it seems to me that the concern about the weak and vulnerable has two aspects. First, there would be a direct concern about weak and vulnerable people in the same unhappy position as Applicants, who do not have the requisite desire (namely "a voluntary, clear, settled and informed decision to commit suicide"), but who either feel that they have some sort of duty to die, or are made to feel (whether intentionally or not) that they have such a duty by family members or others, because their lives are valueless and represent an unjustifiable burden on others. (This aspect is more fully described by Lord Sumption at para 228 below). Secondly, there is a concern that the extension of the law to permit assisted suicide would send a more general message to weak and vulnerable people, who would consequently be more at risk of committing, or seeking assistance to commit, suicide while not having the requisite desire to do so.

The appellants argue that the article 8 rights of Applicants to put an end to their lives, which are rights of a very high order bearing in mind their very cruel circumstances, should not be sacrificed for a merely speculative concern about another class of persons. They say that the harmful effect that liberalising the law on assisting suicide may have on vulnerable and weak people is no more than speculative, because no evidence has been adduced to suggest otherwise, and because in jurisdictions where assisted suicide is permitted, there do not seem to have been any undesirable consequences for the weak and vulnerable.

It is true that the Falconer Report, supported by the reports of the two Canadian panels, states that in the Netherlands, Oregon and Switzerland there is no evidence of abuse of the law, which permits assisting a suicide in prescribed circumstances and subject to conditions. However, negative evidence is often hard to obtain, there is only a limited scope for information given the few jurisdictions where assisted suicide is lawful and the short time for which it has been lawful there, and different countries may have different potential problems. In other words, the evidence on that point plainly falls some way short of establishing that there is no risk. The most that can be said is that the Falconer commission and the Canadian panels could find no evidence of abuse. As Lord Sumption points out in paras 224-225 below, however, while the factual evidence in this connection is sparse, anecdotal, and inconclusive, the expert experienced and professional opinion evidence does provide support for the existence of the risk. In all the circumstances, this concern cannot, in my opinion, possibly be rejected as fanciful or unrealistic.

Having said that, if a proposal were put forward whereby Applicants could be helped to kill themselves, without appreciably endangering the lives of the weak and vulnerable, then this objection could be overcome, or at least circumnavigated. In that connection, Lady Hale, during argument, brought home to me the significance of the point that it has been regarded as quite acceptable in cases such as Re B (Treatment) that the High Court should have the power to accede to a request by an individual that her life support machine be turned off. Furthermore, albeit less relevantly, I note that in the Mental Capacity Act 2005 Parliament has recognised the right of individuals to give advance directions that they be refused medical treatment. In the former case, the appropriate protection for the weak and vulnerable appears to be that a High Court Judge must first be satisfied that the request is based on a settled, informed and voluntary desire. In the latter case, it would seem that a formal document recording the desire will suffice.

The moral arguments

The contention that there is a moral justification for the present law did not feature much in argument, and then only in very general terms. In so far as the argument is based on the sanctity, or primacy, of other human lives, it does little more, in my view, than replicate the concerns about the lives of the weak and vulnerable. In so far as it is based on the sanctity or primacy of Applicants' lives, it has been substantially undermined by the enacting of section 1 of the 1961 Act. I find it hard to see how a life can be said to be sacred if it is lawful for the person whose life it is to end it; to put the point another way, if the primacy of human life does not prevent a person committing suicide, it is difficult to see why it should prevent that person seeking assistance in committing suicide. I also agree with what Lord Wilson says in this connection in paras 199 and 200 below.

Another moral justification briefly advanced for not changing the law was that Parliament did not want to send out a message that human life is to be undervalued. I am somewhat sceptical about semaphore justifications for legislative or judicial decisions, but I accept that we should proceed on the basis this may have some force. However, it seems to me that, once again, this argument is another way of expressing the concern about the need to protect weak and vulnerable people, albeit a larger class of weak and vulnerable people.

There is a rather different moral issue, which was not really covered in argument, namely that, while it is one thing for a person to take his own life, it is another thing to take, or even to assist in the taking of, someone else's life. In other words, there may be a view that, even though it is morally acceptable for people to take their own lives, it would be morally corrupting for another person, and indeed for society as a whole, if that other person could assist people in taking their lives. I think that there would be significantly more force in this point if the assister actually performed the act which caused the death, such as actually administering the barbiturate, as opposed to setting up a system which enables the person who wishes to commit suicide to activate the machine to perform the final act.

In the eyes of the law, there is a very large difference between the two courses: the first is murder or manslaughter, and the second an offence under section 2. In this connection, the decision of the House of Lords in R v Kennedy (No 2) [2008] 1 AC 269 is very much in point. In that case, the House of Lords, in a powerful opinion given by Lord Bingham, overruled a decision that a defendant was guilty of manslaughter when he "had produced a situation in which [the alleged victim] could inject herself [with a lethal drug], in which her self-injection was entirely foreseeable and in which self-injection could not be regarded as extraordinary" on the ground that this decision "conflicted with the rules on personal autonomy and informed voluntary choice" – para 16. Accordingly, "[t]he finding that the deceased freely and voluntarily administered the injection to himself, knowing what it was, is fatal to any contention that the appellant caused the heroin to be administered to the deceased or taken by him" – para 18.

To my mind, the difference between administering the fatal drug to a person and setting up a machine so that the person can administer the drug to himself is not merely a legal distinction. Founded as it is on personal autonomy, I consider that the distinction also sounds in morality. Indeed, authorising a third party to switch off a person's life support machine, as in Bland or Re B (Treatment) seems to me, at least arguably, to be, in some respects, a more drastic interference in that person's life and a more extreme moral step, than authorising a third party to set up a lethal drug delivery system so that a person can, but only if he wishes, activate the system to administer a lethal drug.

Indeed, if one is searching for a satisfactory boundary between euthanasia or mercy killing and assisted suicide, which Lord Sumption discusses at para 227 below, I believe that there may be considerable force in the contention that the answer, both in law and in morality, can best be found by reference to personal autonomy. Subject to those cases where the act can be classified as an "omission" (eg, to my mind somewhat uncomfortably in terms of common sense, switching off a life-supporting machine at least if done by an appropriately authorised person, as in Bland and Re B (Treatment)), it seems to me that if the act which immediately causes the death is that of a third party that may be the wrong side of the line, whereas if the final act is that of the person himself, who carries it out pursuant to a voluntary, clear, settled and informed decision, that is the permissible side of the line. In the latter case, the person concerned has not been "killed" by anyone, but has autonomously exercised his right to end his life. (I should perhaps make it clear that I am not thereby seeking for a moment to cast doubt on the correctness of the decisions in Bland and Re B (Treatment), both of which appear to me to have been plainly rightly decided).

The argument based on the value of human life is not one which can only be raised by the Secretary of State. The evidence shows that, in the light of the current state of the law, some people with a progressive degenerative disease feel themselves forced to end their lives before they would wish to do so, rather than waiting until they are incapable of committing suicide when they need assistance (which would be their preferred option). Section 2 therefore not merely impinges adversely on the personal autonomy of some people with degenerative diseases, but actually, albeit indirectly, may serve to cut short their lives.

For the reasons I have discussed, therefore, while it would be wrong to ignore the moral arguments against permitting Applicants to be assisted to kill themselves, I do not consider that they are particularly telling. Indeed, by requiring one to focus on the important feature of personal autonomy, they appear to me to provide a degree of support for the appellants' case.

In any event, quite apart from the points already made, the mere fact that there are moral issues involved plainly does not mean that the courts have to keep out. Even before the 1998 Act came into force, the courts were prepared to make decisions which developed the law and involved making moral choices of this type. Re B (Wardship), Re J, Bland, Re F and (albeit only by a week) Conjoined Twins were all decided before the 1998 Act was in force, and each decision would have been regarded as involving a wrong moral choice by some people. Further, in Re B (Treatment) the court was prepared to decide that an action should be taken (albeit that it was classified as an omission by Lord Goff) which would end a person's life because that person wanted that action to be taken (although, of course, it should not have been necessary to go to court to give effect to B's wishes, unless there was some concern over her mental capacity or some other special reason). Thus, the courts have been ready both to assume responsibility for developing the law on what are literally life and death issues, and then to shoulder responsibility for implementing the law as so developed. It is perhaps worth noting in the present context that, despite pleas from judges, Parliament has not sought to resolve these questions through statutes, but has been content to leave them to be worked out by the courts.

The argument that the issue should be left to Parliament

The Secretary of State contends that, under our constitutional settlement, the determination of the criminal law on a difficult, sensitive and controversial issue such as assisted suicide is one which is very much for Parliament. There is obvious force in that argument, given that, less than five years ago, Parliament approved the general prohibition on assisting suicide, by redrafting section 2(1), so that it continued to render all cases of assisted suicide criminal, and by leaving subsection (4), with its control by the DPP, in place.

Nonetheless it is self-evident that the mere fact that Parliament has recently enacted or approved a statutory provision does not prevent the courts from holding that it infringes a Convention right. By the 1998 Act, Parliament has cast on the courts the function of deciding whether a statute infringes the Convention. In a case such as the present, where the margin of appreciation applies, a court will only invoke this function where it has concluded that the issue is within its competence, in which case the fact that Parliament has recently considered the issue, while relevant, cannot automatically deprive the courts of their right, indeed their obligation, to consider the issue.

It is not easy to identify in any sort of precise way the location of the boundary between the area where it is legitimate for the courts to step in and rule that a statutory provision, which is not irrational, infringes the Convention and the area where it is not. However, it is not, I think, sensible or even possible to seek to define where the boundary lies. In Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70, the House of Lords had to consider whether, by changing the common law they would be overstepping the boundary which separates legitimate development of the law from judicial legislation. Lord Goff said this at p 173:

"I feel bound however to say that, although I am well aware of the existence of the boundary, I am never quite sure where to find it. Its position seems to vary from case to case. Indeed, if it were to be as firmly and clearly drawn as some of our mentors would wish, I cannot help feeling that a number of leading cases in your Lordships' House would never have been decided the way they were."

If that is the position with regard to a long existing boundary, it is scarcely surprising that it should be the same in relation to a boundary which has been in existence for less than fourteen years.

In connection with the present case, the Secretary of State can justifiably place reliance on Lord Bingham's observations about the Hunting Act 2004 in R (Countryside Alliance) v Attorney General [2008] 1 AC 719, para 45:

"There are of course many … who do not consider that there is a pressing (or any) social need for the ban imposed by the Act. But after an intense debate a majority of the country's democratically elected representatives decided otherwise. It is of course true that the existence of duly enacted legislation does not conclude the issue. … Here we are dealing with a law which is very recent and must … be taken to reflect the conscience of a majority of the nation. The degree of respect to be shown to the considered judgment of a democratic assembly will vary according to the subject matter and the circumstances. But the present case seems to me pre-eminently one in which respect should be shown to what the House of Commons decided. The democratic process is liable to be subverted if, on a question of moral and political judgment, opponents of the Act achieve through the courts what they could not achieve in Parliament."

Those observations serve as a salutary reminder that we, as judges, should be very cautious before being prepared to hold that we should exercise our jurisdiction under section 4 of the 1998 Act in the present case. However, Lord Bingham's words plainly do not by themselves justify a simple refusal to hold that we have or should exercise the jurisdiction. Furthermore, the reasons for, and nature of, the controversy in that case were very different from those in this appeal, and the interference with the article 8 rights of people such as Mr Lamb as a result of section 2 is enormously greater than any arguable alleged interference with the rights of those who wished to hunt in the Countryside Alliance case.

Quite apart from this, there is force in the point that difficult or unpopular decisions which need to be taken, are on some occasions more easily grasped by judges than by the legislature. Although judges are not directly accountable to the electorate, there are occasions when their relative freedom from pressures of the moment enables them to take a more detached view. As Lord Brown said in the Countryside Alliance case at para 158, "[s]ometimes the majority misuses its powers. Not least this may occur when what are perceived as moral issues are involved". However, (save, as some have argued, in circumstances which are very unlikely ever to arise) Parliamentary sovereignty and democratic accountability require that the legislature has the final say, as section 4 of the 1998 Act recognises: Lord Kerr accurately records the position in para 343 below.

As for the other points relied on by the Secretary of State, it is true that in Pretty v DPP the House of Lords unanimously rejected the contention that section 2 infringed the article 8 rights of Ms Pretty, even if such rights were engaged. However, that was immediately after the House had wrongly concluded that her article 8 rights were not engaged, and before the Strasbourg court had considered the issue in the cases referred to in paras 29-38 above. Further, the arguments deployed in Pretty v DPP on this issue were very general in nature (see at p 805D). Indeed, as I shall seek to explain later in this judgment, it seems to me that the arguments deployed by the appellants in this appeal were not sufficiently focussed to justify a declaration of incompatibility in the first appeal.

The extent of the need for assessing views, experiences and expertise, as invoked by the Secretary of State, will depend very much on the nature of the appellants' proposals, as well as the evidence and arguments. Similarly, the degree of familiarity and confidence which the judiciary can claim in relation to the proposal, which will depend on the precise nature of the proposals. However, as the cases considered in paras 21-26 above demonstrate, the courts are used to dealing with life and death issues of the sort to which the present proceedings give rise.

The Secretary of State's reliance on the need for detailed provisions and regulatory safeguards has some force, but the court is not being asked to set up a specific scheme under which Applicants could be assisted to commit suicide such that it would be disproportionate for the law to forbid them from doing so. As Lord Hughes says in para 267 below, it is a matter for Parliament to determine the precise details of any scheme. But that does not prevent the court from concluding that there are a number of possible schemes. For the purpose of deciding that article 8 is infringed, the court needs to consider that aspect no further than is necessary to satisfy itself that some such scheme or schemes could be practically feasible.

It is also relevant to bear in mind the current position, whereby, with Parliament's approval, the policy of the DPP is to investigate any assisted suicide after the event, and to lean against prosecuting where the assister was a close relative or friend activated by compassion, at least where there are no other, aggravating, relevant factors. A system whereby a judge or other independent assessor is satisfied in advance that someone has a voluntary, clear, settled, and informed wish to die and for his suicide then to be organised in an open and professional way, would, at least in my current view, provide greater and more satisfactory protection for the weak and vulnerable, than a system which involves a lawyer from the DPP's office inquiring, after the event, whether the person who had killed himself had such a wish, and also to investigate the actions and motives of any assister, who would, by definition, be emotionally involved and scarcely able to take, or even to have taken, an objective view. It is also appropriate to ask which of those two courses would be more satisfactory for the compassionate friend or relative (whose article 8 rights may also be engaged).

Furthermore, it is clear from the 2010 Policy, the evidence summarised in para 48 above, as well as from the DPP's decision referred to in para 39 above, that those people who, out of compassion, assist relations and friends who wish to commit suicide, by taking or accompanying them to Dignitas, are routinely not prosecuted. In other words, those people who have access to supportive friends and relations, and who possess the means and physical ability to travel to Switzerland, are able in practice to be assisted in their wish to commit suicide, whereas those people, such as Mr Lamb and Martin, who lack one or more of those advantages, cannot receive any such assistance. Further, even those who can in practice be helped to travel to Switzerland to die would, understandably, prefer to die without the upheaval involved, at their homes with dignity in peace.

The point discussed in paras 92-95 above, relating to the moral difference between a doctor administering a lethal injection to an Applicant, and a doctor setting up a lethal injection system which an Applicant can activate himself, is also of significance in relation to institutional competence. It could be said to be a radical step for a court to declare a statutory provision incompatible, if such a declaration involved effectively stating that the law should be changed so as to decriminalise an act which would unquestionably be characterised as murder or (if there were appropriately mitigating circumstances) manslaughter. If, on the other hand, Dr Nitschke's machine, described in para 4 above, could be used, then a declaration of incompatibility would be a less radical proposition for a court to contemplate.

Conclusion on this issue

In my view, bearing in mind all the features discussed in the preceding 26 paragraphs, the arguments raised by the Secretary of State do not justify this Court ruling out the possibility that it could make a declaration of incompatibility in relation to section 2. The interference with Applicants' article 8 rights is grave, the arguments in favour of the current law are by no means overwhelming, the present official attitude to assisted suicide seems in practice to come close to tolerating it in certain situations, the appeal raises issues similar to those which the courts have determined under the common law, the rational connection between the aim and effect of section 2 is fairly weak, and no compelling reason has been made out for the court simply ceding any jurisdiction to Parliament.

Accordingly, while I respect and understand the contrary opinion, so well articulated by Lord Sumption and Lord Hughes, I am of the view that, provided that the evidence and the arguments justified such a conclusion, we could properly hold that that section 2 infringed article 8. A court would therefore have to consider an application to make a declaration of incompatibility on its merits, and it seems to me that it would be inappropriate for us to fetter the judiciary's role in this connection in advance. More specifically, where the court has jurisdiction on an issue falling within the margin of appreciation, I think it would be wrong in principle to rule out exercising that jurisdiction if Parliament addresses the issue: it could be said with force that such an approach would be an abdication of judicial responsibility. In that connection, I agree with what Lord Mance says in para 191 below. Further, in practical terms, given the potential for rapid changes in moral values and medicine, it seems to me that such an approach may well turn out to be inappropriate in relation to this particular issue.

However, I consider that, even if it would otherwise be right to do so on the evidence and arguments which have been raised on the first appeal, it would not be appropriate to grant a declaration of incompatibility at this time. In my opinion, before making such a declaration, we should accord Parliament the opportunity of considering whether to amend section 2 so as to enable Applicants, and, quite possibly others, to be assisted in ending their lives, subject of course to such regulations and other protective features as Parliament thinks appropriate, in the light of what may be said to be the provisional views of this Court, as set out in our judgments in these appeals.

It would, of course, be unusual for a court to hold that a statutory provision, conventionally construed, infringed a Convention right and could not be construed compatibly with it, and yet to refuse to make a declaration under section 4 of the 1998 Act. However, there can be no doubt that there is such a power: section 4(2) states that if there is an incompatibility, the court "may" make a declaration to that effect, and the power to grant declaratory relief is anyway inherently discretionary. The possibility of not granting a declaration of incompatibility to enable the legislature to consider the position is by no means a novel notion. As pointed out by Lady Hale, Lord Nicholls in Bellinger v Bellinger [2003] 2 AC 467, para 53, said this:

"It may also be that there are circumstances where maintaining an offending law in operation for a reasonable period pending enactment of corrective legislation is justifiable. An individual may not then be able, during the transitional period, to complain that his rights have been violated. The admissibility decision of the court in Walden v Liechtenstein (Application no 33916/96) (unreported) 16 March 2000 is an example of this pragmatic approach to the practicalities of government."

In my view, even if the facts and arguments otherwise justified a declaration of incompatibility on the first appeal (which for the reasons given below, I consider they do not), this is one of those exceptional cases where it would have been inappropriate to grant a declaration of incompatibility at this stage. That view is based on considerations of proportionality in the context of institutional competence and legitimacy which are well articulated by Lord Mance in paras 166-170 below, taking forward his discussion in Re G, referred to in paras 71-73 above.

There is a number of reasons which, when taken together, persuade me that it would be institutionally inappropriate at this juncture for a court to declare that section 2 is incompatible with article 8, as opposed to giving Parliament the opportunity to consider the position without a declaration. First, the question whether the provisions of section 2 should be modified raises a difficult, controversial and sensitive issue, with moral and religious dimensions, which undoubtedly justifies a relatively cautious approach from the courts. Secondly, this is not a case like Re G where the incompatibility is simple to identify and simple to cure: whether, and if so how, to amend section 2 would require much anxious consideration from the legislature; this also suggests that the courts should, as it were, take matters relatively slowly. Thirdly, section 2 has, as mentioned above, been considered on a number of occasions in Parliament, and it is currently due to be debated in the House of Lords in the near future; so this is a case where the legislature is and has been actively considering the issue. Fourthly, less than thirteen years ago, the House of Lords in Pretty v DPP gave Parliament to understand that a declaration of incompatibility in relation to section 2 would be inappropriate, a view reinforced by the conclusions reached by the Divisional Court and the Court of Appeal in this case: a declaration of incompatibility on this appeal would represent an unheralded volte-face.

In para 204 below, Lord Wilson refers to the power of the court under section 4 of the 1998 Act as giving rise to mechanism for "collaboration" between the courts and Parliament, and many judges and academics have referred to the "dialogue" which takes place between national courts and the Strasbourg court. While those expressions should not detract from the seriousness of a declaration of incompatibility, they may be helpful metaphors. Dialogue or collaboration, whether formal or informal, can be carried on with varying degrees of emphasis or firmness, and there are times when an indication, rather than firm words are more appropriate and can reasonably be expected to carry more credibility. For the reasons just given, I would have concluded that this was such a case.

Parliament now has the opportunity to address the issue of whether section 2 should be relaxed or modified, and if so how, in the knowledge that, if it is not satisfactorily addressed, there is a real prospect that a further, and successful, application for a declaration of incompatibility may be made. It would not be appropriate or even possible to identify in advance what amounts to a reasonable time in this context. However, bearing in mind the predicament of the Applicants, and the attention the matter has been given inside and outside Parliament over the past twelve years, one would expect to see the issue whether there should be any and if so what legislation covering those in the situation of Applicants explicitly debated in the near future, either along with, or in addition to, the question whether there should be legislation along the lines of Lord Falconer's proposals. Nor would it be possible or appropriate to identify in advance what would constitute satisfactory addressing of the issue, or what would follow once Parliament had debated the issue: that is something which would have to be judged if and when a further application is made, as indicated in para 112 above. So that there is no misunderstanding, I should add that it may transpire that, even if Parliament did not amend section 2, there should still be no declaration of incompatibility: that is a matter which can only be decided if and when another application is brought for such a declaration. In that connection, Lord Wilson's list of factors in para 205 below, while of real interest, might fairly be said to be somewhat premature.

Should the Court grant a declaration of incompatibility? – issue (d)

This question does not need to be answered in the light of the conclusion I have reached in the immediately preceding paragraphs. However, it would, I think, be wrong to leave the first appeal without stating that, even if I had concluded that it would in principle have been institutionally appropriate to make a declaration of incompatibility in these proceedings, I would not have done so on the basis of the evidence and arguments laid before the courts.

Before we could uphold the contention that section 2 infringed the article 8 rights of Applicants, we would in my view have to have been satisfied that there was a physically and administratively feasible and robust system whereby Applicants could be assisted to kill themselves, and that the reasonable concerns expressed by the Secretary of State (particularly the concern to protect the weak and vulnerable) were sufficiently met so as to render the absolute ban on suicide disproportionate. I do not consider that we can be properly confident that we have the evidence or that the courts below or the Secretary of State have had a proper opportunity to address the issue, in order to determine whether requirement (c) or (d) in Aguilar Quila is satisfied.

That brings me to the appellants' specific proposals, which in my view suffered from a lack of proper focus. As I understand it, they rely heavily on the recommendations of the Falconer Report and the conclusions of Smith J in the Canadian case of Carter v Canada [2012] BCSC 886, but I would find it hard to accept either of them as a sound basis for supporting the appellants' case. So far as the Carter case is concerned, I have nothing to add to what Lord Mance says at paras 178-182 below.

As for the Falconer Commission, in common with the proponents of change in 2006 and 2009 in the House of Lords, it recommended that section 2 should be cut down only to the extent that assistance could be accorded to those who were terminally ill – with twelve months or less to live. (I believe that Lord Falconer is currently proposing a shorter period, six months.) That would not assist Applicants. Further, I find it a somewhat unsatisfactory suggestion. Quite apart from the notorious difficulty in assessing life expectancy even for the terminally ill, there seems to me to be significantly more justification in assisting people to die if they have the prospect of living for many years a life that they regarded as valueless, miserable and often painful, than if they have only a few months left to live.

Further, the Falconer Report suggests that the decision whether to permit someone to be assisted to die should be left to doctors. That is understandable (though I am not entirely convinced by it) if the issue is whether the person concerned will die shortly. However, if the people who are to be assisted are in the sad situation of Applicants, I would have thought that there is much to be said for the idea, first mooted by Lady Hale and developed in her judgment in paras 314-316 below, that it should be a High Court Judge who decides the issue. Indeed, it appears to me that it may well be that the risks to the weak and vulnerable could be eliminated or reduced to an acceptable level, if no assistance could be given to a person who wishes to die unless and until a Judge of the High Court has been satisfied that his wish to do so was voluntary, clear, settled and informed.

As explained in paras 21-26 above, over the past twenty-five years, the High Court has been able to sanction a number of actions in relation to people which will lead to their deaths or will represent serious invasions of their body – sterilisation, denial of treatment, withdrawal of artificial nutrition and hydration, switching off a life support machine, and surgery causing death to preserve the life of another. It is true that in most of these cases, the court is involved because the person concerned cannot express his wishes. However, that is not true of cases such as Re B (Treatment), where the issue for the court would be identical to that in the type of case raised by the appellants.

In these circumstances, I consider that it is certainly conceivable that a court could conclude that section 2 infringes article 8 in so far as it precludes an Applicant from receiving assistance in committing suicide, provided that a High Court Judge has formally determined that he has a voluntary, clear, settled and informed wish to do so.

However, over and above the reason discussed in paras 113-118 above, it would not have been appropriate to reach such a conclusion in these proceedings. Neither the Secretary of State nor the courts below have had a proper opportunity to consider this, or any other, proposal. As Lord Mance explains more fully in paras 175-177 below, in both the High Court and the Court of Appeal, the claim of a declaration of incompatibility was rather a fall-back argument, and the appellants contended that the issue could not be determined without further fact-finding. Further, the argument in those courts was primarily advanced on the basis that someone would actually have to kill Mr Nicklinson and Mr Lamb, as opposed to enabling them to administer a fatal dose themselves through operating an eyeblink computer, and, for the reasons given in paras 92-95 and 110 above, the ability of an Applicant to commit suicide through the use of a machine such as the eye blink computer is of importance in my view.

In any event, at least on the basis of the arguments and evidence which have been put before the Court, there would have been too many uncertainties to justify our making a declaration of incompatibility. Of course, it is for Parliament to decide how to respond to a declaration of incompatibility, and in particular how to change the law. However, at least in a case such as this, the Court would owe a duty, not least to Parliament, not to grant a declaration without having reached and expressed some idea of how the incompatibility identified by the court could be remedied.

Thus, it appears to me that it would be necessary to consider purely factual matters, such as whether devices such as Dr Nitschke's machine are reliable, whether they could be activated by Applicants, and whether it would be feasible to use them. There would also be mixed factual and policy issues to consider, such as whether appropriate safeguards (including by whom and on what basis the decision to permit an assisted suicide should be made) could be developed to protect both those who firmly wish to die and those who do not, whether Applicants could be fairly identified and regulated as a self-contained collection of people, whether there would be implications for people who were not Applicants but wished to be assisted in killing themselves, and if so what the implications were, and how they should be dealt with.

The disposal of the first appeal – issue (e)

In these circumstances, I consider that we should dismiss the first appeal.

However, it is right to add that, if I had concluded that article 8 was infringed by section 2 as conventionally interpreted, I would have had no hesitation in rejecting the appellants' contention that section 2 could be read, in the light of section 3 of the 1998 Act, so as to comply with the Convention. The only argument put forward to support the contention was that, a person who assisted an Applicant to die could rely on the doctrine of necessity to avoid criminal liability under section 2. As Lord Dyson and Elias LJ explained in para 25 of their judgment in the Court of Appeal, to extend the defence of necessity to a charge of assisted suicide would be a revolutionary step, which would be wholly inconsistent with both recent judicial dicta of high authority, and the legislature's intentions. As to judicial dicta, see R v Howe [1987] 1 AC 417, 429B-D and 453B-F, per Lord Hailsham and Lord Mackay respectively, Bland, pp 892E-893A per Lord Mustill, and Inglis at para 37, per Lord Judge CJ. So far as legislative intention is concerned, in 1961, Parliament decided, through section 2(1), to create a statutory offence of assisting a suicide in a provision which admitted of no exceptions, and it confirmed that decision as recently as 2009 (when section 2(1) was repealed and re-enacted in more detailed terms) following a debate in which the possibility of relaxing the law on the topic was specifically debated.

I turn then to the issues raised by the DPP's appeal.

Does the 2010 Policy infringe article 8? – issue (f)

The challenge to the validity of the 2010 Policy

In Purdy at para 41, Lord Hope explained that any law which restricts a Convention right must satisfy the two requirements of accessibility and foreseeability. He went on to explain that the requirement of foreseeability is satisfied where the person concerned "is able to foresee....the consequences which a given action may entail", a formulation which was derived from the Sunday Times case, para 49, and a number of subsequent decisions of the Strasbourg court. "The level of precision required of domestic legislation", as was stated in Hasan and Chaush, para 84, "depends to a considerable degree on the content of the instrument in question, the field it is designed to cover and the number and status of those to whom it is addressed".

The decision in Purdy was not merely justified by the fact that the crime of assisting suicide can engage articles 2 and 8; it was more because the crime, at least in many cases, has a unique combination of features, all of which point firmly towards a requirement for clear guidance. First, section 2(1) renders it a crime to assist someone else to do an act which is not itself in any way a crime. Secondly the victim is not merely a willing participant, but the instigator. Thirdly, the victim's article 8 rights are interfered with unless the crime is committed. Fourthly, the person committing the offence will be a reluctant participant, motivated by compassion for the so-called victim, and not by emotions which normally stimulate criminal behaviour.

It is true that the last three of these four characteristics are not an inevitable feature of a case of assisting a suicide, but they will all frequently feature in such cases. Indeed, it was because assisting suicide was such an unusual crime that subsection (4) was included in section 2 – see Purdy, para 46. Even more centrally, it was because all four characteristics were such likely features of a potential offence under section 2 that Purdy was decided in the way that it was. The requirement for a specific policy was not to protect the interests of those who were contemplating putting pressure on the vulnerable, or seeking to benefit from someone's suicide, but to protect the interests of the very people assistance of whose suicide would involve all four characteristics – see Purdy, paras 53, 68, 86 and 102.

The need for a clear policy in this area is said to be supported by the reasoning of the majority of the Strasbourg court in Gross. It concerned a somewhat different aspect of assisted dying, but the court's emphasis in para 66 on the need for guidelines to avoid a person being "in a state of anguish and uncertainty regarding the extent of her right to end her life", seems to me to apply to a case such as those that have given rise to these appeals. I note also the conclusion in para 69 that it was "up to the domestic authorities to issue comprehensive and clear legal guidelines" as to "whether and under which circumstances an individual … not suffering from a terminal illness should be granted the ability to acquire a legal dose of medication allowing them to end their life".

Martin's argument in the second appeal is that, as a result of a lack of clarity in the 2010 Policy, the law relating to the crime of assisting suicide fails to live up to the foreseeability requirement. The lack of clarity is said to arise where a person who has a voluntary, clear, settled and informed wish to die and who requires assistance, is given such assistance by a third party, who is acting purely out of compassion and who has exerted no pressure on the person, but is not a relation or friend, and would often be a doctor or other professional carer. Where the third party is a friend or relation, then in the absence of any aggravating factor, the 2010 Policy indicates that a prosecution would be unlikely, but in any other case the position could fairly be described as more opaque.

The evidence suggests that this uncertain state of affairs leads doctors and other professional carers almost always to refuse to give any information or advice to those who wish to end their lives. This degree of caution, although understandable, appears to go too far, and I gladly associate myself with the accurate and helpful guidance given in para 255(2), (3) and (4) of Lord Sumption's judgment.

Having said that, Lord Dyson MR and Elias LJ expressed the problem which was said to exist with the 2010 Policy very well at para 140 of the Court of Appeal's judgment:

"How does [the 2010 Policy] apply in the case of a medical doctor or nurse who is caring for a patient and out of compassion is willing to assist the patient to commit suicide, but is not, as it were, in the business of assisting individuals to commit suicide and perhaps has never done so before? How much weight is given by the DPP to para 43(14) alone? And if the professional accepts some payment for undertaking the task, will that be likely to involve a finding that he or she is not wholly motivated by compassion, thereby triggering both paragraph 43(6) and paragraph 43(13)? These questions are of crucial importance to healthcare professionals who may be contemplating providing assistance. It is of no less importance to victims who wish to commit suicide, but have no relative or close friend who is willing and able to help them to do so. Suppose that (i) none of the factors set out in para 43 is present (apart from the para 43(14) factor) and (ii) all of the factors set out in para 44 are present. What is the likelihood of a prosecution in such a situation? The Policy does not say. To adopt the language of the Sunday Times case, even in such a situation, the Policy does not enable the healthcare professional to foresee to a reasonable degree the consequences of providing assistance. … In short, we accept the submission … that the Policy does not provide medical doctors and other professionals with the kind of steer … that it provides to relatives and close friends acting out of compassion …."

Is it appropriate to expect greater foreseeability?

Lord Hughes and Lord Kerr rightly point out that (i) the state of the law is clear, indeed could not be clearer, in the sense that any form of assisting a suicide is a crime under the unconditional provisions of section 2, and (ii) the role of the DPP is constitutionally limited, in that it is not, and indeed cannot be, to make the law, let alone to change the law, but to decide how much guidance she can properly give in her policy with regard to prosecutions under section 2(1). We are not therefore in the same area as that which was being discussed in the passages cited from Gross in para 135 above, which was concerned with what conduct would be lawful in Swiss law. Further, any policy which the DPP has (whether published or not) must be applied after the event. In these circumstances, it is inevitable that any policy issued by the DPP has to retain a degree of flexibility: each case has to be assessed after the event by reference to its own particular facts.

However, I do not share Lord Hughes's concerns about (i) the decision in Purdy (the correctness of which was not challenged by anyone in these appeals), or (ii) the risk of a spill-over into other statutory crimes where there is a provision such as section 2(4). As to (i), particularly given the unique combination of features identified in para 133 above, it was appropriate to require the DPP to publish a policy in relation to assisting suicide, given that his existing general code did not satisfactorily apply to that crime. It was not as if the House was seeking to say what that policy should be. As to (ii), although section 2(4) was given weight in Purdy, it is the DPP's general prosecutorial discretion which is the relevant power which gave rise to the decision in that case. More importantly, as already mentioned, it is the unique character of the offence, coupled with the decision in Pretty v UK, which led the House to decide that a specific published policy for assisting suicide was required.

Accordingly, we are here concerned with a very unusual crime which is the subject of a specific policy. However, that does not undermine the force of the constitutional argument that it is one thing for the court to decide that the DPP must publish a policy, and quite another for the court to dictate what should be in that policy. The purpose of the DPP publishing a code or policy is not to enable those who wish to commit a crime to know in advance whether they will get away with it. It is to ensure that, as far as is possible in practice and appropriate in principle, the DPP's policy is publicly available so that everyone knows what it is, and can see whether it is being applied consistently. While many may regret the fact that the DPP's policy is not clearer than it is in relation to assistance given by people who are neither family members nor close friends of the victim, and while many may believe that the policy should be the same for some categories of people who are not family members or close friends as for those who are, it would not be right for a court in effect to dictate to the DPP what her policy should be.

A further point

In these circumstances, were it not for one point, I would simply have accepted the DPP's case on the second appeal. However, the matter is not quite so simple in light of what was said by Lord Judge CJ (dissenting on this point) in the Court of Appeal about the 2010 Policy:

"185. … [I]t seems clear to me that paragraph 14 addresses the risks which can arise when someone in a position of authority or trust, and on whom the victim would therefore depend to a greater or lesser extent, assisting in the suicide in circumstances in which, just because of the position of authority and trust, the person in authority might be able to exercise undue influence over the victim. As I read this paragraph it does not extend to an individual who happens to be a member of a profession, or indeed a professional carer, brought in from outside, without previous influence or authority over the victim, or his family, for the simple purposes of assisting the suicide after the victim has reached his or her own settled decision to end life, when, although emotionally supportive of him, his wife cannot provide the necessary physical assistance.

186. … Naturally, it would come as no surprise at all for the DPP to decide that a prosecution would be inappropriate in a situation where a loving spouse or partner, as a final act of devotion and compassion assisted the suicide of an individual who had made a clear, final and settled termination to end his or her own life. The Policy … deliberately does not restrict the decision to withhold consent to family members or close friends acting out of love and devotion. The Policy certainly does not lead to what would otherwise be an extraordinary anomaly, that those who are brought in to help from outside the family circle … are more likely to be prosecuted than a family member when they do no more than replace a loving member of the family, acting out of compassion, who supports the 'victim' to achieve his desired suicide. The stranger brought into this situation, who is not profiteering, but rather assisting to provide services which, if provided by the wife, would not attract a prosecution, seems to me most unlikely to be prosecuted. In my respectful judgment this Policy is sufficiently clear to enable Martin, or anyone who assists him, to make an informed decision about the likelihood of prosecution."

For the reasons given by Lord Dyson MR and Elias LJ quoted in para 138 above, I do not agree with Lord Judge CJ that one can spell out of the 2010 Policy the approach which he sets out so clearly in those two paragraphs. However, the important point for present purposes is that what is said in those two paragraphs represents, according to her counsel on instructions, the view of the DPP herself, as to the appropriate policy. If the DPP's policy does not mean what she intends it to mean, and this has been made clear in open court, then it is her duty, both as a matter of domestic public law and in the light of the Strasbourg jurisprudence as a "public authority", to ensure that the confusion is resolved.

However, I am of the view that it would not be appropriate, at least at this stage, to make an order which would require the DPP to amend the 2010 Policy. Rather, I think, it is appropriate to leave it to her to review the terms of the 2010 Policy, after consultation if she thinks fit, with a view to amending it so as to reflect the concerns expressed in the judgments of this Court, and any other concerns which she considers it appropriate to accommodate.

There are three reasons which persuade me that it would be inappropriate to make any order against the DPP at this stage. First, it is really only as a result of the hearing of this appeal that it has become clear that the 2010 Policy may not reflect the DPP's views. It would therefore be somewhat harsh for the court to impose a duty on her to deal with the problem, as opposed to giving her the opportunity to do so. Secondly, although her agreement with Lord Judge CJ's analysis was no doubt considered, the DPP should not be regarded as bound by it. She should have a proper opportunity to consider the 2010 Policy, after making such enquiries as she thinks appropriate. Thirdly, in any event, the contents of any order would either be very vague or they would risk doing that which the court should not do, namely usurping the functions of the DPP, or even of Parliament.

Given that, in an important respect, the 2010 Policy does not appear to reflect what the DPP intends, it seems to me inevitable that she will take appropriate steps to deal with the problem, particularly in the light of the impressive way in which her predecessor reacted to the decision in Purdy. However, if the confusion is not sorted out, then, at least in my view, the court's powers could be properly invoked to require appropriate action, but, as I have said, it seems very unlikely that this will be necessary.

The contents of the Policy – issue (g)

In the light of my conclusion in the immediately preceding paragraphs, Martin's cross-appeal does not arise.

Conclusions

For the reasons I have given (which are generally the same as those of Lord Mance) I would summarise my conclusions as follows:

a) In common with all other members of the Court, I do not consider that section 2 imposes what the Strasbourg court would regard as an impermissible "blanket ban" on assisted suicide, which would take it outside the margin of appreciation afforded on this issue to member states;

b) Given that the Strasbourg court has decided that it is for the member states to decide whether their own law on assisted suicide infringes article 8, I consider, in common with other members of the Court, that domestic courts have the constitutional competence to decide the issue whether section 2 infringes article 8;

c) (i) Unlike Lord Sumption, Lord Clarke, Lord Reed and Lord Hughes, I do not consider that it would be institutionally inappropriate, or only institutionally appropriate if Parliament refuses to address the issue, for a domestic court to consider whether section 2 infringes the Convention, but,

(ii) Unlike Lady Hale and Lord Kerr, I do not consider that it would be institutionally appropriate for us to determine the issue at this time;

d) Notwithstanding the views of Lady Hale and Lord Kerr to the contrary, I am of the view that, quite apart from my view in para (c)(ii), in the light of the evidence and the arguments presented on this appeal the Court is not in a position to decide the issue;

e) In common with all members of the Court, I do not consider that the Court should involve itself with the terms of the DPP's policy on assisted suicide, albeit that I would expect the DPP to clarify her policy.

In these circumstances, I would dismiss the appeal brought by Mrs Nicklinson and Mr Lamb, allow the appeal brought by the DPP, and dismiss the cross-appeal brought by Martin.

LORD MANCE

I agree generally with the reasoning and conclusions of Lord Neuberger on the appeals by Mrs Nicklinson and Mr Lamb, read with the following observations of my own. On the appeal and cross-appeal in the case of Martin, I agree that the Director of Public Prosecution's appeal should be allowed and Martin's cross-appeal dismissed, for reasons given by Lord Neuberger and Lord Sumption, supplemented by short observations of my own.

The appeals by Mrs Nicklinson and Mr Lamb

Before us the appeals by Mrs Nicklinson and Mr Lamb have acquired a different focus from that of Mr Nicklinson's case below. Below, Mr Nicklinson's case, as recorded by Toulson LJ in paras 15 and 21 of his judgment in the Divisional Court, was that "the only way in which [he] could end his life other than by self-starvation would be by voluntary euthanasia". Although a statement had been produced by a North Australian doctor, Dr Nitschke, to the effect that it would be technologically possible for Mr Nicklinson "to take the final step of initiating suicide with the aid of a machine which Dr Nitschke has invented", pre-loaded with lethal drugs and capable of being digitally activated by Mr Nicklinson by a blink of his eye (para 16), Toulson LJ went on to say that "In these circumstances [Mr Nicklinson] wants to be able to choose to end his life by voluntary euthanasia" at a moment of his choosing (para 17); and he added that, although Dr Nitschke's evidence meant that the claim that s.2 of the Suicide Act 1961 was incompatible with article 8 of the Convention" was "not entirely academic", the main part of the argument on Mr Nicklinson's behalf under article 8 was directed to establishing that it "requires voluntary active euthanasia to be permitted by law" (para 21).

In the Divisional Court (para 122) and Court of Appeal (para 105), the cases of R (Pretty) v Director of Public Prosecutions [2001] UKHL 61, [2002] 1 AC 800 and Pretty v United Kingdom (2002) 35 EHRR 1 were treated as binding on the issue whether the blanket ban contained in s.2 of the Suicide Act is compatible with the Convention as interpreted by the Strasbourg court. The Court of Appeal added the caveat that the court must also satisfy itself as to the proportionality of the ban as a matter of domestic law (para 110), but concluded that "in a case like this, it would be improper for a court to find a blanket prohibition disproportionate where this is not dictated by Strasbourg jurisprudence" (para 111).

In the courts below, therefore, the main focus was on Mr Nicklinson's submissions that necessity should be recognised as a defence to murder at common law and/or in the light of article 8 of the European Convention on Human Rights. That case is not now pursued. The case now advanced is that a machine like Dr Nitschke's would offer a feasible means of suicide, and that the prohibition on assisting suicide in s.2(1) of the Suicide Act 1961, as amended by s.59 of the Coroners and Justice Act 2009, should be read down to permit this assistance to be volunteered, or if that is not possible that the prohibition should be declared incompatible with article 8 of the Convention on Human Rights.

In my opinion the decision of the European Court of Human Rights in Pretty v United Kingdom (2002) 35 EHRR 1 establishes at the international level that it is within the margin of appreciation of Member States of the Council of Europe to legislate in terms involving a blanket prohibition of assisted suicide. More recent cases, such as Haas v Switzerland (2011) 53 EHRR 33, Koch v Germany (2013) 56 EHRR 6 and Gross v Switzerland (2014) 58 EHRR 7 throw no doubt on this, since they concern either a state (Switzerland) which permits assisted suicide or a state (Germany) whose courts had acted contrary to article 6 of the Convention by refusing even to address the issue.

It is of interest to compare the European Court of Human Rights decision in Pretty with the majority reasoning of the United States Supreme Court in Washington v Glucksberg 521 U.S. 702 (1997). The United States Supreme Court was concerned with the due process clause in the American Constitution, under which a wide range of fundamental liberties has in the past been recognised, including the right to marry, to have and direct the upbringing of children and to have an abortion (Roe v Wade 410 U.S. 113 (1973) and Planned Parenthood v Casey 505 U.S. 833 (1992) and the right to refuse unwanted lifesaving medical treatment (Cruzen v Director, Missouri Dept. of Health 497 U.S. 261 (1990). It held that the right to due process did not extend to a right to commit assisted suicide, and that the State of Washington's blanket prohibition on assisted suicide was accordingly not unconstitutional. It noted that the overwhelming majority of States prohibited assisted suicide, some after quite recent debates about it, but it also noted that voters in Oregon had in 1994 enacted a "Death with Dignity Act" legalising physician-assisted suicide for competent, terminally ill adults (p 717). Its comment was that this showed that "the States are currently engaged in serious, thoughtful examination of physician-assisted suicide and other similar issues" (p 719). It is a comment of some relevance in my opinion to the position in which this Court finds itself in relation to Parliament, a subject to which I shall return.

I do not read paragraph 76 of the European Court of Human Right's judgment in Pretty as suggesting that a blanket prohibition may be incompatible with article 8 at the international level. I agree with Lord Neuberger's analysis in his paragraphs 62 to 65. When the European Court of Human Rights said in paragraph 76 in Pretty that:

"It does not appear to be arbitrary to the Court for the law to reflect the importance of the right to life, by prohibiting assisted suicide while providing for a system of enforcement and adjudication which allows due regard to be given in each particular case to the public interest in bringing a prosecution, as well as to the fair and proper requirements of retribution and deterrence."

it was, as I see it, reaffirming the legitimacy (at the international level and bearing in mind the margin of appreciation) of a blanket prohibition, but recognising that, at the subsequent stage of enforcement and adjudication, some flexibility in approach was appropriate.

In Purdy, at para 74, Lord Brown thought it implicit in the Court's reasoning in Pretty v United Kingdom

"that in certain cases, not merely will it be appropriate not to prosecute, but a prosecution under section 2(1) would actually be inappropriate".

He went on:

"If in practice the ban were to operate on a blanket basis, the only relaxation in its impact being by way of merciful sentences on some occasions when it is disobeyed, that would hardly give sufficient weight to the article 8 rights with which the ban, if obeyed, is acknowledged to interfere."

The emphasis in this passage is on the distinction between merciful sentencing and the decisions not to prosecute at all which the Director is expressly authorised to take under s.2(4). The passage does not suggest that a blanket ban is in principle impermissible (if it did, it would be contrary to much else that the Court said in Pretty v United Kingdom and later cases). It is recognising the exercise of the Director's discretion under s.2(4) as an important concomitant of the blanket ban in the United Kingdom context. But it is a concomitant, not intended to undermine or qualify the legitimacy of the blanket prohibition, but directed to the treatment of those who infringe it.

In Haas, para 55, the Court observed that "the vast majority of member States seem to attach more weight to the protection of the individual's life than to his or her right to terminate it. It follows that the States enjoy a considerable margin of appreciation in this area." In Koch v Germany (2013) 56 EHRR 6, para 70, the Court repeated its reference to a "considerable" margin of appreciation. It is, in these circumstances, important to note how the Court put the position under article 8 in Haas at para 51, and repeated it in Koch, para 52 and Gross, para 59. It said that, in the light of the previous case-law:

"an individual's right to decide by what means and at what point his or her life will end, provided he or she is capable of freely reaching a decision on this question and acting in consequence, is one of the aspects of the right to respect for private life within the meaning of article 8 of the Convention."

It would be wrong in my view to deduce from this that the Strasbourg jurisprudence accepts that those capable of freely reaching a decision to end their lives, but physically incapable of bringing that about by themselves, have a prima facie right to obtain voluntary assistance, which is now the issue in this case, to achieve their wish. article 8.1 is, on the authority of Pretty v United Kingdom, engaged in this area. But it does not by itself create a right. A right only exists (at least in any coherent sense) if and when it is concluded under article 8.2 that there is no justification for a ban or restriction.

Autonomy is an important value. But, as soon as the giving of assistance to those physically incapable of committing suicide without assistance comes into question, other factors, in particular the wider implications for third parties (not just the voluntary assister), also require consideration. The European Court of Human Rights' words "capable of …. acting in consequence" were carefully devised.

To distinguish in this respect between those capable of committing suicide by themselves and others is not unjustifiably to discriminate against the latter. A submission to contrary effect was rejected by the House of Lords in R (Pretty), where Lord Bingham said:

"She contends that the section is discriminatory because it prevents the disabled, but not the able bodied, exercising their right to commit suicide. This argument is in my opinion based on a misconception. The law confers no right to commit suicide."

A similar answer was also given by the European Court of Human Rights in Pretty v United Kingdom. In relation to the applicant's complaint "that she has been discriminated against in the enjoyment of the rights guaranteed under that provision in that domestic law permits able-bodied persons to commit suicide yet prevents an incapacitated person from receiving assistance in committing suicide" (para 86), the Court said:

"87. For the purposes of article 14 a difference in treatment between persons in analogous or relevantly similar positions is discriminatory if it has no objective and reasonable justification, that is if it does not pursue a legitimate aim or if there is not a reasonable relationship of proportionality between the means employed and the aim sought to be realised. Moreover, the Contracting States enjoy a margin of appreciation in assessing whether and to what extent differences in otherwise similar situations justify a different treatment (see Camp and Bourimi v. the Netherlands, no. 28369/95, § 37, ECHR 2000-X). Discrimination may also arise where States without an objective and reasonable justification fail to treat differently persons whose situations are significantly different (see Thlimmenos v. Greece [GC], no. 34369/97, § 44, ECHR 2000-IV).

88. Even if the principle derived from Thlimmenos was applied to the applicant's situation however, there is, in the Court's view, objective and reasonable justification for not distinguishing in law between those who are and those who are not physically capable of committing suicide. Under article 8 of the Convention, the Court has found that there are sound reasons for not introducing into the law exceptions to cater for those who are deemed not to be vulnerable (see paragraph 74 above). Similar cogent reasons exist under article 14 for not seeking to distinguish between those who are able and those who are unable to commit suicide unaided. The borderline between the two categories will often be a very fine one and to seek to build into the law an exemption for those judged to be incapable of committing suicide would seriously undermine the protection of life which the 1961 Act was intended to safeguard and greatly increase the risk of abuse."

It follows from the margin of appreciation which exists at the international level that it is for domestic courts to examine the merits of any claim to receive assistance to commit suicide: see Koch, para 71. The United Kingdom position is on the face of it clear. Parliament has legislated for a blanket prohibition, combined with a discretion on the part of the Director of Public Prosecutors to decide whether in any particular case to prosecute. Pursuant to the House of Lords' decision in R (Purdy) v Director of Public Prosecutions [2010] 1 AC 345, the Director has issued his 2010 Policy statement, set out in Lord Neuberger's judgment at paras 46 and 47.

As the Court of Appeal noted (para 110), the fact that Parliament has legislated a blanket ban is not the end of the matter as far as United Kingdom courts are concerned. Under the Human Rights Act 1998, it is the courts' role to consider United Kingdom legislation in the light of the Convention rights scheduled to that Act. Where a "considerable" margin of appreciation exists at the international level, both the legislature and the judiciary have a potential role in assessing whether the law is at the domestic level compatible with such rights. That means considering whether a blanket prohibition is in accordance with law, in the sense that it not only meets a legitimate aim, but does so in a way which is necessary and proportionate. The legislator's choice is not necessarily the end of the matter: see In re G (Adoption: Unmarried Couple) [2008] UKHL 38; [2009] 1 AC 173.

At this point, however, questions of institutional competence arise at the domestic level. The interpretation and ambit of s.2 are on their face clear and general, and whether they should be read down or declared incompatible in the light of article 8 raises difficult and sensitive issues. Context is all, and these may well be issues with which a court is less well equipped and Parliament is better equipped to address than is the case with other, more familiar issues. On some issues, personal liberty and access to justice being prime examples, the judiciary can claim greater expertise than it can on some others. The same applies to the legislature – even though I fully accept, that, while the legislature is there to reflect the democratic will of the majority, the judiciary is there to protect minority interests, and to ensure the fair and equal treatment of all. Whether a statutory prohibition is proportionate is, in my view, a question in the answering of which it may well be appropriate to give very significant weight to the judgments and choices arrived at by the legislator, particularly when dealing with primary legislation.

In their impressive judgments in the courts below, Toulson LJ (at paras 57 to 62 and 75 to 84) and Lord Dyson MR (at paras 49 and 56 to 60) cited extensively from prior authority cautioning against courts' interference in difficult ethical and social issues better fitted for Parliamentary resolution under our democratic traditions. One such case was Airedale NGHS Trust v. Bland [1992] UKHL 5; [1993] AC 789, where the House of Lords addressed the narrow but vital distinction between mercy killing and the discontinuance of life-sustaining measures in the context of an application to discontinue measures of the latter kind in respect of a patient in a permanent vegetative state. In this context, Lord Browne-Wilkinson said (p.880A-B) that

"it is not for the judges to seek to develop new, all embracing, principles of law in a way which reflects the individual judges' moral stance when society as a whole is substantially divided on the relevant moral issues" (p 880A-B per Lord Brown-Wilkinson

Lord Mustill said (p.890G-891C):

"These are only fragments of a much wider nest of questions, all entirely ethical in content, beginning with the most general- "Is it ever right to terminate the life of a patient, with or without his consent?" I believe that adversarial proceedings, even with the help of an amicus curiae, are not the right vehicle for the discussion of this broad and highly contentious moral issue, nor do I believe that the judges are best fitted to carry it out. On the latter aspect I would adopt the very blunt words of Scalia J. in Cruzan v. Director, Missouri Department of Health (1990) 110 S.Ct. 2841, 2859, where a very similar problem arose in a different constitutional and legal framework. These are problems properly decided by the citizens, through their elected representatives, not by the courts.

My Lords, I believe that I have said enough to explain why, from the outset, I have felt serious doubts about whether this question is justiciable, not in the technical sense, but in the sense of being a proper subject for legal adjudication. The whole matter cries out for exploration in depth by Parliament and then for the establishment by legislation not only of a new set of ethically and intellectually consistent rules, distinct from the general criminal law, but also of a sound procedural framework within which the rules can be applied to individual cases. The rapid advance of medical technology makes this an ever more urgent task, and I venture to hope that Parliament will soon take it in hand. Meanwhile, the present case cannot wait. We must ascertain the current state of the law and see whether it can be reconciled with the conduct which the doctors propose."

In that case, as Lord Mustill's final sentences indicate, the House had to address the point under the law as it then stood. I note however that the United States Supreme Court reached a similar result in another decision under the due process clause: Vacco v Quill 521 U.S. 793 (1997), handed down on the same day as Washington v Glucksberg. Rejecting an argument that the State of New York's ban on assisted suicide by the prescription of lethal medication to mentally competent, terminally ill patients suffering great pain was unconstitutional, the Supreme Court said that "the distinction between assisting suicide and withdrawing life-sustaining treatment, a distinction widely recognised and endorsed in the medical profession and in our legal traditions, is both important and logical; it is certainly rational" (pp 800-801), and that even though "the line between the two may not be clear, …. certainty is not required, even were it possible" (p 808). In the present appeal, current United Kingdom law is clear. Prior to the Human Rights Act 1998 that would have been the end of the matter. The question is how far the Human Rights Act requires a different approach.

It is in my view a mistake to approach proportionality as a test under the Human Rights Act which is insensitive to considerations of institutional competence and legitimacy. The qualifying objectives reflected in article 8.2 of the Convention can engage responsibilities normally attaching in the first instance to other branches of the state, whether the executive or the legislature. When considering whether a particular measure is necessary and all the more when considering whether it is justified on a balancing of competing and often incommensurate interests, courts should recognise that there can still be wisdom and relevance in the factors mentioned in the preceding two paragraphs. This is all the more so when the court is considering the scope of the Convention rights, as enacted domestically, in a situation, like the present, which the European Court of Human Rights has held to fall within the United Kingdom's international margin of appreciation. That Parliament has regularly addressed the general area and is still actively engaged in considering associated issues in the context of Lord Falconer's Assisted Dying Bill 2013 underlines the significance of the point. This does not mean that there is a legal rule that courts will not intervene (as to which see Lord Steyn, extra-judicially in Deference: A Tangled Story, [2005] PL 345, commenting on R (ProLife Alliance) v British Broadcasting Corp [2003] UKHL 23, [2004] 1 AC 185, paras 74-77 per Lord Hoffmann) or that the courts have no role. It means merely that some judgments on issues such as the comparative acceptability of differing disadvantages, risks and benefits have to be and are made by those other branches of the state in the performance of their everyday roles, and that courts cannot and should not act, and do not have the competence to act, as a primary decision-maker in every situation. Proportionality should in this respect be seen as a flexible doctrine.

That institutional competence is important in the context of judgments made on issues of proportionality has been recognised in a series of cases: see e.g. A v Secretary of State for the Home Department [2004] UKHL 56; [2005] 2 AC 68, paras 29 and 38-39, per Lord Bingham, R (SB) v Governors of Denbigh High School [2006] UKHL 15, [2007] 1 AC 100, para 34, per Lord Bingham, R (Animal Defenders International) v Secretary of State for Culture, Media and Sport [2008] UKHL 15, [2008] 1 AC 1312, para 53, per Lady Hale, R (Countryside Alliance) v Attorney General [2007] UKHL 52, [2008] 1 AC 719, para 45, per Lord Bingham (the passage quoted by Lord Neuberger in his para 102) and Bank Mellat v HM Treasury (No 2) [2013] UKSC 39, [2013] 3 WLR 179, paras 68-76 per Lord Reed, with whose observations in these paragraphs Lord Sumption, Lady Hale, Lord Kerr and Lord Clarke agreed at para 20 and Lord Neuberger agreed at para 166.

Lord Reed's observations, worth study in their entirety, included the following:

"71. An assessment of proportionality inevitably involves a value judgment at the stage at which a balance has to be struck between the importance of the objective pursued and the value of the right intruded upon. The principle does not however entitle the courts simply to substitute their own assessment for that of the decision-maker. As I have noted, the intensity of review under EU law and the Convention varies according to the nature of the right at stake and the context in which the interference occurs. Those are not however the only relevant factors. One important factor in relation to the Convention is that the Strasbourg court recognises that it may be less well placed than a national court to decide whether an appropriate balance has been struck in the particular national context. For that reason, in the Convention case law the principle of proportionality is indissolubly linked to the concept of the margin of appreciation. That concept does not apply in the same way at the national level, where the degree of restraint practised by courts in applying the principle of proportionality, and the extent to which they will respect the judgment of the primary decision maker, will depend on the context, and will in part reflect national traditions and institutional culture. For these reasons, the approach adopted to proportionality at the national level cannot simply mirror that of the Strasbourg court.

….

74. The judgment of Dickson CJ in Oakes (R v Oakes [1986] 1 SCR 103) provides the clearest and most influential judicial analysis of proportionality within the common law tradition of legal reasoning. Its attraction as a heuristic tool is that, by breaking down an assessment of proportionality into distinct elements, it can clarify different aspects of such an assessment, and make value judgments more explicit. The approach adopted in Oakes can be summarised by saying that it is necessary to determine (1) whether the objective of the measure is sufficiently important to justify the limitation of a protected right, (2) whether the measure is rationally connected to the objective, (3) whether a less intrusive measure could have been used without unacceptably compromising the achievement of the objective, and (4) whether, balancing the severity of the measure's effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, the former outweighs the latter. The first three of these are the criteria listed by Lord Clyde in de Freitas (de Freitas v Permanent Secretary if Ministry of Agriculture, Fisheries, Lands and Housing [1999] 1 AC 69), and the fourth reflects the additional observation made in Huang (Huang v Secretary of State for the Home Department [2007] 2 AC 167). I have formulated the fourth criterion in greater detail than Lord Sumption JSC, but there is no difference of substance. In essence, the question at step four is whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure.

75. In relation to the third of these criteria, Dickson CJ made clear in R v Edwards Books and Art Ltd [1986] 2 SCR 713, 781-782 that the limitation of the protected right must be one that 'it was reasonable for the legislature to impose', and that the courts were 'not called upon to substitute judicial opinions for legislative ones as to the place at which to draw a precise line'. This approach is unavoidable, if there is to be any real prospect of a limitation on rights being justified: as Blackmun J once observed, a judge would be unimaginative indeed if he could not come up with something a little less drastic or a little less restrictive in almost any situation, and thereby enable himself to vote to strike legislation down (Illinois State Board of Elections v Socialist Workers Party (1979) 440 US 173, 188–189); especially, one might add, if he is unaware of the relevant practicalities and indifferent to considerations of cost. To allow the legislature a margin of appreciation is also essential if a federal system such as that of Canada, or a devolved system such as that of the United Kingdom, is to work, since a strict application of a 'least restrictive means' test would allow only one legislative response to an objective that involved limiting a protected right.

76. In relation to the fourth criterion, there is a meaningful distinction to be drawn (as was explained by McLachlin CJ in Alberta v Hutterian Brethren of Wilson Colony [2009] 2 SCR 567, para 76) between the question whether a particular objective is in principle sufficiently important to justify limiting a particular right (step one), and the question whether, having determined that no less drastic means of achieving the objective are available, the impact of the rights infringement is disproportionate to the likely benefits of the impugned measure (step four)."

As Lord Reed also observed at para 69:

"69. Proportionality has become one of the general principles of EU law, and appears in article 5(4) of the EU Treaty. The test is expressed in more compressed and general terms than in German or Canadian law, and the relevant jurisprudence is not always clear, at least to a reader from a common law tradition. In R v Ministry of Agriculture, Fisheries and Food, Ex p Fedesa (Case C-331/88) [1990] ECR I-4023, the European Court of Justice stated, at para 13):

"The court has consistently held that the principle of proportionality is one of the general principles of Community law. By virtue of that principle, the lawfulness of the prohibition of an economic activity is subject to the condition that the prohibitory measures are appropriate and necessary in order to achieve the objectives legitimately pursued by the legislation in question; when there is a choice between several appropriate measures recourse must be had to the least onerous, and the disadvantages caused must not be disproportionate to the aims pursued."

The intensity with which the test is applied – that is to say, the degree of weight or respect given to the assessment of the primary decision-maker - depends on the context."

The flexibility of proportionality in the parallel context of European Union law was underlined in the Court of Appeal with regard to legislative choices made by a minister in R (Sinclair Collis Ltd) v Secretary of State for Health [2011] EWCA Civ 437, [2012] QB 394 (see especially at paras 126-134 and 203 per Arden LJ and Lord Neuberger MR, respectively) and was, still more recently, underlined in my judgment (in which Lord Neuberger and Lord Clarke joined) in Kennedy v The Charity Commission [2014] UKSC 20, para 54. It is also demonstrated instructively in the context of Convention law in an article by Julian Rivers, Proportionality and Variable Intensity of Review (2006) 65 CLJ 174.

The main justification advanced for an absolute prohibition on assisting suicide, even in cases as tragic as Mr Nicklinson's and Mr Lamb's, is the perceived risk to the lives of other, vulnerable individuals who might feel themselves a burden to their family, friends or society and might, if assisted suicide were permitted, be persuaded or convince themselves that they should undertake it, when they would not otherwise do so. The relevant measure is the prohibition, which on this basis has a legitimate aim. Whether it is rationally connected to that aim depends upon the existence of the perceived risk. Whether it is necessary depends upon whether a lesser measure would have achieved, or at least not "unacceptably" have compromised, the aim. Whether it is proportionate depends upon identifying what the measure achieves and balancing this against the consequences for other interests. These four stages, derived from the passage in Lord Reed's judgment in Bank Mellat quoted in para 168 above, are analytically useful. They are also subject to some modification in particular contexts, not here directly relevant. (For example, the third stage may not apply in quite the same way under article 1 of Protocol No 1.) The third and fourth stages may raise potentially overlapping considerations, but the distinction between them is important. The third asks whether the aim could have been achieved without significant compromise by some less intrusive measure. The fourth involves the critical exercise of balancing the advantages of achieving the aim in the way chosen by the measure against the disadvantages to other interests. This balancing exercise, often involving the weighing of quite different rights or interests, is a core feature of the court's role, and can be described as involving proportionality in the strict sense of that word. How intensely the court will undertake the exercise, and to what extent the court will attach weight to the judgment of the primary decision-maker (be it legislature or executive), depends at each stage on the context, in particular the nature of the measure and of the respective rights or interests involved. The primary decision maker's choices as to the aim to adopt and the measure to achieve it may be entitled to considerable respect. But at the fourth stage other interests may come into play, the intrinsic and comparative weight of which the court may be as well or even better placed to judge in the light of all the material put before it.

The existence of a risk to other vulnerable individuals is a premise of the decisions of the European Court of Human Rights at the international level. Thus, in Pretty v United Kingdom, para 74, the Court said of s.2(1) of the Suicide Act 1961:

"Doubtless the condition of terminally ill individuals will vary. But many will be vulnerable and it is the vulnerability of the class which provides the rationale for the law in question. It is primarily for States to assess the risk and the likely incidence of abuse if the general prohibition on assisted suicides were relaxed or if exceptions were to be created. Clear risks of abuse do exist, notwithstanding arguments as to the possibility of safeguards and protective procedures."

Further, at the United Kingdom domestic level, the existence of such a risk was also accepted by the House of Lords in R (Pretty) as an alternative ground of decision if article 8.1 was engaged.

It is submitted on the present appeal that developments and further evidence available since the Pretty v United Kingdom and Purdy cases require the Supreme Court to reach a conclusion opposite to its considered view in R (Pretty) (which also formed the starting point for its decision in Purdy) that the blanket prohibition in s.2(1) was proportionate. As to this, first, I do not consider that either the reasoning on legal or other issues or the decision in Pretty v United Kingdom and the more recent Strasbourg cases of Haas, Koch and Gross or the House's reasoning or decision in Purdy affect the view expressed on this point in R (Pretty). The experience acquired regarding the s.2(4) discretion does not mean that the principle needs reconsideration. Of 85 cases referred to the CPS between 1 April 2009 and 1 October 2013, 64 were not proceeded with and 11 were withdrawn. 9 are ongoing and only 1 has been successfully prosecuted. The Director's discretion is evidently effective to avoid prosecutions which would serve no useful purpose after the event, but these figures do not appear to me to bear on the appropriateness of the blanket prohibition or on risks that could develop without it.

I would accept that it is in principle open to claimants in the position of the appellants to invite a court to revisit an issue of proportionality previously decided between different parties in the light of different evidence, and, further, that this would not involve inviting the Supreme Court to depart precedentially from Purdy. Proportionality is here a judgment reached in the light of evidence, so that it is capable of being re-litigated in this way, although courts should no doubt discourage such re-litigation in the absence of fresh and significantly different evidence.

However, examination of the course of the present case raises in my view serious questions about its suitability for any such exercise. At no stage does this litigation appear to have been approached on the basis that the court should hear primary evidence about the issues. There has been nothing like the wide-ranging examination of expert and statistical material concerning suicide and the psychological factors and risks bearing on its occurrence which appears to have informed the United States Supreme Court's judgments in Washington v Glucksberg. Much of the material put before the Supreme Court on the present appeal has been second hand, adduced in other litigation or by other inquiries. Thus Toulson LJ, when referring to the January 2012 report of the Commission on Assisted Dying chaired by Lord Falconer, said (para 24):

"We were asked to read the report and have done so. However, it is important to stress that it was not an officially appointed commission. Its report contains an interesting analysis of arguments and views, but it would not be right for the court to treat it as having some form of official or quasi-official status".

The report in fact records that "some prominent individuals and organisations that are fundamentally opposed to any form of assisted dying being legally permitted in the UK" refused to participate in giving evidence (p.39). Toulson LJ also records (para 25) that after judgment was given at first instance by Smith J in Carter v Canada [2012] BCSC 886, counsel for Mr Nicklinson applied for leave to introduce the evidence in that case into the present case, recognising "that, if it were admitted, there would have to be a further hearing in order to enable the witnesses to be called and cross-examined".

Similarly, before the Court of Appeal counsel accepted that determination of the question whether there had been a disproportionate interference with article 8 rights would involve "consideration of a vast array of detailed evidence, including sociological, philosophical and medical material, which would have to be conducted by the Divisional Court". Before the Supreme Court, on the other hand, the appellants' primary case has become not to "invite the Supreme Court to embark upon a close study of the evidence that is now available of the relative risks and advantages of relaxing the prohibitions on assisted suicide", but instead to submit that the Supreme Court "can strike the necessary balance without such a forensic exercise because it has been conducted already by a number of expert bodies whose conclusions are remarkably similar and upon whose conclusions the Court can place weight". In the alternative, "if the Court considers that it cannot carry out the balancing exercise without further exploration of the underlying evidential issues", they repeat their request that the case should be remitted to the High Court for that exercise to be conducted along the lines of that in Carter v Canada [2012] BCSC 886, with appropriate guidance as to how the balancing exercise is to be conducted.

The appellants' primary case before the Supreme Court amounts in substance to an invitation to short-cut potentially sensitive and difficult issues of fact and expertise, by relying on secondary material. There can in my opinion be no question of doing that. Their secondary case (their primary case below) is that the case should in effect re-commence from the beginning with directions for evidence to be called and examined on the relevant issues of fact. But the handing down of the first instance decision in Carter v Canada shortly before the Divisional Court hearing is not a justification for not applying at the outset for a trial of the relevant issues on the basis of evidence directly examined before the court.

The main basis relied upon for departing from the view expressed in Pretty is the fresh evidence said to have been gained in the meantime. That comes, first, from those few states where assisted suicide is lawful (Switzerland, Oregon, Vermont and Montana) or where both euthanasia and assisted suicide are lawful (the Netherlands, Belgium and Luxembourg), and, second, from other sources, such as the Falconer Commission on Assisted Dying (January 2012), the Royal Society of Canada ("RSC") Expert Panel on End-of-Life Decision Making (2011), the Quebec "Dying with Dignity" Select Committee Report (March 2012) and the examination of the issue by Smith J at first instance in Carter v Canada [2012] BCSC 886 (over-ruled at [2013] BCCA 435 on the ground that the issue was covered by the prior authority of Rodriguez v British Columbia (Attorney General) [1993] 3 SCR 519). As I have already noted, the Falconer Commission did not receive (though it would have liked to) the evidence of committed opponents of the idea of assisted suicide, and some of the other evidence is open to the comment that it was commissioned by or involved persons already on record as committed to a change in the law.

In Carter v Canada, where both claimants suffered from intractable and progressive diseases, the RSC Report was also tendered without there being the opportunity to cross-examine its makers. The Government of Canada criticised it as "essentially argument on one side of the debate (and largely legal argument), rather than a balanced or comprehensive review of the issues", and noted that three of the authors were expert witnesses for the plaintiffs, while a fourth author had been assisting the plaintiffs with instructing expert witnesses. Canada also called another Fellow of the Royal Society of Canada, who said that, in his view, "the RSC Report reads as though it was written with a pre-ordained conclusion", commented on the rapidity with which the panel had proceeded, and noted that its membership lacked representation from the palliative care community, and included persons who had previously expressed views supportive of physician-assisted dying.

In the event, Smith J said this:

"[129] I have now reviewed the RSC Report and have concluded that it will be admitted in evidence, in the main for the fact that the expert panel made the recommendations that it did. I have not relied upon it as evidence on any contentious matters such as the efficacy of safeguards in jurisdictions that permit physician-assisted dying. Its review of the legal landscape regarding end-of-life care in Canada is not evidence, but the equivalent of a law review article or a legal text."

In Rodriguez every judge at every level had agreed that the purpose of protecting vulnerable persons from inducement to commit suicide was pressing and substantial, and it was also held that the prohibition on assisted suicide was rationally connected to that purpose. No challenge was made to either conclusion in Carter v Canada. The issue there was focused on whether the prohibition was the minimum step necessary and was proportionate in the pursuit of that purpose. Smith J said that considerable deference was due to Parliament on that issue, but that this did not relieve the court of its role in assessing such matters. Ultimately, she concluded:

"1267. With respect to the absolute prohibition's alleged salutary effects in preventing wrongful deaths, or in preventing abuse of vulnerable people, my review of the evidence from Canada and elsewhere leaves me unconvinced that an absolute prohibition has that effect in comparison with a prohibition combined with stringently limited exceptions."

On that basis, she concluded: "the benefits of the impugned law are not worth the costs of the rights limitations they create" (para 1285).

It is in my view clear from the judgment at first instance in Carter v Canada and from even the superficial examination of the evidence which the appellants now in effect invite as their primary case (paragraph 175 above) that it would be impossible for this Court to arrive at any reliable conclusion about the validity of any risks involved in relaxing the absolute prohibition on assisting suicide, or (which is surely another side of the same coin) the nature or reliability of any safeguards which might accompany and make possible such a relaxation, without detailed examination of first-hand evidence, accompanied by cross-examination. This has not occurred in this case, but, in its absence, I do not see how one can accept the appellants' submission that the circumstances have so changed that R (Pretty) v Director of Public Prosecutions should now no longer be followed.

Whatever else may be said about the evidential position, it is not in my opinion sustainable to suggest that there is no evidence and to describe as ruminations a conclusion that permitting assisted suicide in the case of persons in Mr Nicklinson's and Mr Lamb's position would pose a relevant risk to vulnerable people (compare paras 349 to 351 of Lord Kerr's opinion). There is a rational connection between the current prohibition in s.2(1) and its aim. As I have already mentioned, both R (Pretty) and Pretty v United Kingdom proceed on that basis. So too, the United States Supreme Court in Renquist CJ's forceful majority judgment in Washington v Glucksberg regarded it as "unquestionable" that the State of Washington's ban on assisted suicide was "rationally related to legitimate government interests" (p 728). I also note that Lord Joffé himself, when moving the second reading of his Assisted Dying for the Terminally Ill Bill on 12 May 2006 (Hansard, col 1188) said:

"When I gave evidence to the Select Committee about the original Bill, I expressed my personal conviction, which was honestly held at the time, that I would welcome a widening of the scope of the legislation. I no longer hold that view. One of the advantages of the Select Committee process was the opportunity to see different regimes in operation, and to hear a wealth of evidence from those who have thought deeply about the issues and are intimately involved in them. At the end of the process, it is now my firm view that the extent of legislative change that I put before the House today …. will have the most advantage and carry the least risk. I would not support further extension into the field of euthanasia, or support assisted dying for patients who are not terminally ill. Others, of course, may have different views, but after three years of legislative effort on the subject, I have no intention of pursuing this issue beyond the ambit of the present Bill".

The Falconer Commission also concluded that it could only recommend a relaxation of s.2 of the Suicide Act in respect of the terminally ill, and Lord Falconer's bill, like Lord Joffe's bill was so confined (though the End of Life Assistance (Scotland) Bill introduced in Scotland in January 2010 and defeated by 18 votes to 16 in December 2010 covered persons (a) diagnosed as terminally ill and finding life intolerable or (b) permanently physically incapacitated to such an extent as not to be able to live independently and finding life intolerable). The Falconer Commission heard evidence about and accepted the risks of any greater extension. It said in its summary of its conclusions at p.27:

"The Commission accepts that there is a real risk that some individuals might come under pressure to request an assisted death if this option should become available, including direct pressures from family members or medical professionals, indirect pressures caused by societal discrimination or lack of availability of resources for care and support, and self-imposed pressures that could result from the individuals having low self-worth or feeling themselves to be a burden on others".

Giving a specific example, the Falconer Commission recorded at p.201 the evidence of Professor Raymond Tallis representing Healthcare Professionals for Assisted Dying, who cautioned against any such extension, with the words:

"I think that there are genuine dangers in extending the scope of assisted dying to people who are not terminally ill, who are disabled. All those things that disability groups fear, I think that it would certainly play into those appropriate fears".

The Falconer Commission also

"received evidence from many disabled people and does not consider that it would be acceptable to recommend that a non-terminally ill person with significant physical impairments should be made eligible under any future legislation to request assistance in ending his or her life" (p.27).

Finally, the Falconer Commission reported (p.323):

"The Commission was unable to reach a consensus on the issue of whether a person who has had a catastrophically life-changing event that has caused them to be profoundly incapacitated should be able to request an assisted death, and we consider that this lack of consensus reflects the mixed views of society on this issue. Bearing in mind the considerable concerns of many disabled people about such a provision, we have recommended that it would not be appropriate for such a provision to be included in future legislation."

The most persuasive case that may be made on behalf of persons in the tragic positions of Mr Nicklinson or Mr Lamb is that they represent a distinct and relatively small group, within which it should be possible to identify in advance by a careful prior review (possibly involving the court as well as medical opinion) those capable of forming a free and informed decision to commit suicide and distinguish them from those who might be vulnerable; and that, on this basis, any risks associated with other groups, or with any proposal that might be made to allow assisted suicide within other groups, can and should be disregarded. On such a basis, it may be argued that the current blanket prohibition is unnecessary or disproportionate. The present position is that some persons (whether or not capable of committing suicide unaided) are assisted to do so (unlawfully though it be) without any such prior review. Further, decisions such as Bland to which I have referred in para 165 above and the further cases referred to by Lord Neuberger in paras 21 to 26 and 98 show that the law and courts are already deeply engaged in issues of life and death. Lord Neuberger also shows in paras 92 to 97 that assisting a suicide could be seen not only as promoting the autonomy of the person committing suicide, but also as involving a less drastic interference in life than some interferences already authorised by law, and conceivably also as enabling some people to postpone suicide. A system permitting assisted suicide in limited circumstances such as the present after careful prior review could on its face have some positive benefits - when compared with the current blanket prohibition, coupled with the de facto occurrence of assisted suicides in relation to which the Director of Public Prosecution has to undertake the more difficult task after the event of deciding whether the suicide assisted was the result of "a voluntary, clear, settled and informed decision": see the Director's guideline number 1 tending to weigh against, and guideline number 4 weighing in favour of, prosecution.

The case which I have outlined in the previous paragraph in favour of a relaxing of the prohibition on assisted suicide is not however one on which even the Falconer Commission was able to reach agreement, and it would at the very least require detailed expert investigation and evidence before its premises could be accepted. This is so, quite apart from any argument that it would be difficult if not impossible to determine what should be the ambit of the persons who should be entitled to take advantage of any relaxation of the current prohibition – difficult in particular to draw the line between the sort of unbearable suffering which persons in the position of Mr Nicklinson and Mr Lamb undergo and the suffering which others not subject to their physical disability may subjectively feel (which would in turn raise the question what is meant by unbearable suffering, touched on by the Falconer Commission at pp.202-203.)

Toulson LJ (at paras 85 to 86) observed correctly that the courts could not themselves fashion any scheme which would define circumstances in which or safeguards subject to which assisted suicide might be appropriate. By the same token, it is impossible, at least on present material, to say with confidence in advance that any such scheme could satisfactorily and appropriately be fashioned. This militates strongly against the courts intervening in this area, at least at this stage, to declare s.2 incompatible at the domestic level, when it is compatible at the international level. In saying this, I note that the Joint Committee on Human Rights in its Seventh Report of Session 2002-2003 (HL Paper 74, HC 547) and Twelfth Report of Session 2003-2004 (HL Paper 93, HC 603) was in each case generally content with the safeguards proposed in respect of assisted suicide of the terminally ill in Lord Joffé's bill. But the terminally ill represent a different group which may call for different safeguards from those which the present would require; the current focus of legislative proposals on the terminally ill may also be influenced by the thought that, since their life expectancy is short, the consequences of any risks materialising of the sort identified by the Falconer Commission at p 27 (para 185 above) may be seen as less serious.

Moreover, any assessment of evidence about risks and potential safeguards must inevitably raise questions regarding the degree of residual risk which is acceptable in this present context. In Carter v Canada (paras 1196-1199) the Government of Canada's argument, that the legislation was justified because its purpose was to eliminate all risk, was, not surprisingly, rejected. But any relaxation of the present blanket prohibition would require value judgments of difficulty and delicacy – in particular, how much risk would attach to and be acceptable in consequence of a relaxation coupled with the introduction of safeguards, and how such risk should be measured against the benefits to persons such as Mr Nicklinson and Mr Lamb, in relation to whom it may be said with certainty that they formed their wish to commit suicide with clear and independent minds, so that there was and is no such risk. The issue at this point is primarily how to assess and balance the factors bearing on acceptability and proportionality which arise for consideration at the third and/or fourth stages of the exercise identified by Lord Reed in Bank Mellat, para 74. As in the different context of Sinclair Collis, so too here I think that the legislator's assessment of the value of the evidence and of the choices to be made in its light is entitled to considerable weight, even if the evidence appears to a court weaker and less conclusive than it might be: see e.g. Sinclair Collis, paras 161, per Arden LJ, and 236-239 and 255, per Lord Neuberger MR.

In these circumstances, the position has not been shown by any convincing evidence to have changed materially since R (Pretty) v Director of Public Prosecutions, and I would refuse to make a declaration of incompatibility. In the light of the way in which it has been presented and pursued, remission to the Divisional Court would not be appropriate. To remit would in reality amount to ordering the case to begin over again with a fresh first instance investigation involving a full examination of expert evidence. I see no basis for that exceptional course. I am also influenced in the view that this is not an appropriate time to contemplate such an investigation by, firstly, the very frequent consideration that Parliament has given to the subject over recent years (see Lord Neuberger's judgment, para 51) and by, secondly, the knowledge that Parliament currently has before it the Assisted Dying Bill and the hope that this may also give Parliament an opportunity to consider the plight of individuals in the position of Mr Nicklinson and Mr Lamb. Parliament has to date taken a clear stance, but this will give Parliament the opportunity to confirm, alter or develop its position. I would, in particular, associate myself at this point with Lord Neuberger's conclusions at paras 110 to 117 of his judgment. While I would, like him, not rule out the future possibility of a further application, I would, as matters presently stand, adapt to the present context a thought which Renquist CJ expressed in a slightly different context in Washington v Glucksberg, p 735: that there is currently "an earnest and profound debate about the morality, legality, and practicality of …. assisted suicide" and "[o]ur holding permits this debate to continue, as it should in a democratic society". Parliament is certainly the preferable forum in which any decision should be made, after full investigation and consideration, in a manner which will command popular acceptance.

However, (and as is implicit in paras 164 et seq above) this does not mean that I agree with Lord Sumption's view that it would be unconstitutional for the courts to consider in the present context whether Parliament's ultimate decision meets the requirements of the Convention rights scheduled to the Human Rights Act 1998, or that, in considering this, the courts' role is limited to assessing the "rationality" of Parliament's decision, as I understand that paras 230, 233 and 234 of Lord Sumption's judgment may suggest. Ultimately, Parliament has itself assigned to the courts a constitutional role in balancing the relevant interests, public and private. Lord Sumption accepts that, in performing this role, courts may - "up to a point" - be required to confront the moral consequences of their decisions (para 233). But, although judges must work within a framework of legal principle, reasoning and precedent, very little, if any, judicial decision-making, especially at an appellate level, is or ought to be separated from a consideration of what is just or fair, and the balancing of interests required under the Human Rights Convention merely underlines this. In circumstances such as the present, it may be incumbent on a court to weigh social risks to the wider public and the moral convictions of a body of members of the public together with values of human autonomy and of human dignity in life and death advocated by other members, and in doing so it will attach great significance to the judgment of the democratically informed legislature. But Lord Sumption's view that that legislative judgment must, in the present social and moral context, necessarily be determinative, reminds me of a submission raised by the Attorney General and rejected by the House of Lords in a political context in A v Secretary of State for the Home Department [2005] 2 AC 68, paras 37-42, where Lord Bingham said (para 42):

"I do not in particular accept the distinction which [the Attorney General] drew between democratic institutions and the courts. It is of course true that the judges in this country are not elected and are not answerable to Parliament. It is also of course true, as pointed out in para 29 above, that Parliament, the executive and the courts have different functions. But the function of independent judges charged to interpret and apply the law is universally recognised as a cardinal feature of the modern democratic state, a cornerstone of the rule of law itself. The Attorney General is fully entitled to insist on the proper limits of judicial authority, but he is wrong to stigmatise judicial decision-making as in some way undemocratic. It is particularly inappropriate in a case such as the present in which Parliament has expressly legislated in section 6 of the 1998 Act to render unlawful any act of a public authority, including a court, incompatible with a Convention right, has required courts (in section 2) to take account of relevant Strasbourg jurisprudence, has (in section 3) required courts, so far as possible, to give effect to Convention rights and has conferred a right of appeal on derogation issues. The effect is not, of course, to override the sovereign legislative authority of the Queen in Parliament, since if primary legislation is declared to be incompatible the validity of the legislation is unaffected (section 4(6)) and the remedy lies with the appropriate minister (section 10), who is answerable to Parliament. The 1998 Act gives the courts a very specific, wholly democratic, mandate. As Professor Jowell has put it

'The courts are charged by Parliament with delineating the boundaries of a rights-based democracy' ('Judicial Deference: servility, civility or institutional capacity?" [2003] PL 592, 597)'."

The appeal and cross-appeal in Martin's case.

In Purdy the House held the Director was under a duty to clarify his position as to the factors which he regarded as relevant for and against prosecution in such a case and was required to promulgate an offence-specific policy identifying the facts and circumstances which he would take into account in deciding whether a prosecution should be brought. The criticism made of the current 2010 policy is that factor (14) recited as favouring prosecution (viz, that the suspect was acting in his or her capacity as a healthcare professional and the victim was in his or her care) leaves unclear the Director's policy in a case where such a professional, without previous influence or authority over the person proposing to commit suicide, renders assistance to that end in the period immediately before the suicide, motivated by compassion. In the Court of Appeal, Lord Judge CJ, in an interpretation which the Director expressly endorsed before the Supreme Court, explained (para 185) that factor (14) was not intended to embrace healthcare professionals "brought in from outside, without previous influence or authority over the victim, or his family, for the simply purpose of assisting the suicide after the victim has reached his or her own settled decision to end life".

I agree with both Lord Neuberger and Lord Sumption that it is not clear that factor (14) has this significance. But I would not order the Director to clarify it in the sense explained by the Director. As Lord Sumption observes, it is open to question whether the sense confirmed by the Director before the Supreme Court would on consideration prove to be consistent with other aspects of the Director's policy, particularly those arising from factors (6), (12), (13) and (16) set out as favouring prosecution. I agree with Lord Neuberger and Lord Sumption that there is nothing on the face of the policy as it presently stands which is open to objection, and that the only appropriate course, in the light of the discussion and submissions before this Court, is that the Director should be left to consider the position and either confirm or reformulate her policy, as she may then decide.

I have considered Lord Sumption's summary of the current legal position in his para 255. The second sentence of para (1) of that summary may be open to different interpretations, and I have stated my own approach to s.2(1) of the Suicide Act in this judgment. In all other respects, I find useful and agree with Lord Sumption's summary.

I would therefore allow the Director's appeal and dismiss Martin's cross-appeal. I would leave her to review her published policy in the light of the judgments given on this appeal, and to confirm or reformulate it as she may or may not then decide to be appropriate.

LORD WILSON

At the end of the six months in which all the members of this court have deliberated upon these appeals with an intensity unique in my experience, I find myself in agreement with the judgment of Lord Neuberger.

I regard his crucial conclusions on the first appeal as the following:

(a) The evidence before the court is not such as to enable it to declare that section 2(1) of the 1961 Act either was incompatible with the rights of Mr Nicklinson or is incompatible with the rights of Mr Lamb (para 119).

(b) For the evidence does not enable the court to be satisfied either that there is a feasible and robust system whereby those in their position can be assisted to commit suicide or that the reasonable concerns of the Secretary of State, particularly to protect the weak and vulnerable, can be sufficiently met so as to render the absolute ban in the subsection disproportionate (para 120).

(c) Even were the evidence such as to have enabled the court to make it, a declaration of incompatibility would at this stage have been inappropriate (para 115).

(d) It would have been inappropriate because, even prior to the making of any declaration, Parliament should have the opportunity to consider whether, and if so how, to amend the subsection to permit assistance to commit suicide to be given to those in the position of Mr Nicklinson and Mr Lamb (para 116).

(e) In particular because the Assisted Dying Bill is presently before it, it would be reasonable to expect Parliament in the near future to enlarge its consideration so as to encompass the impact of the subsection on those in their position (para 118).

(f) Were Parliament not satisfactorily to address that issue, there is a real prospect that a further, and successful, application for a declaration of incompatibility might be made (para 118).

(g) The risks to the weak and vulnerable might well be eliminated, or reduced to an acceptable level, were Parliament to provide that assistance might be given to those in their position only after a judge of the High Court had been satisfied that their wish to commit suicide was voluntary, clear, settled and informed (para 123).

Lady Hale and Lord Kerr put forward a powerful case for making a declaration of incompatibility even at this stage. But two principal objections are levelled against it.

The first objection is founded upon the sanctity (or, for those for whom that word has no meaning, the supreme value) of life which, for obvious reasons, is hard-wired into the minds of every living person. It lies at the heart of the common law and of international human rights and it is also an ethical principle of the first magnitude. As Hoffmann LJ suggested in his classic judgment in the Court of Appeal in Airedale NHS Trust v Bland [1993] AC 789 at 826, a law will forfeit necessary support if it pays no attention to the ethical dimension of its decisions. In para 209 below Lord Sumption quotes Hoffmann LJ's articulation of that principle but it is worth remembering that Hoffmann LJ then proceeded to identify two other ethical principles, namely those of individual autonomy and of respect for human dignity, which can run the other way. In the Pretty case, at para 65, the ECtHR was later to describe those principles as of the very essence of the ECHR. It was in the light (among other things) of the force of those two principles that in the Bland case the House of Lords ruled that it was lawful in certain circumstances for a doctor not to continue to provide life-sustaining treatment to a person in a persistent vegetative state but – relevantly to the practical resolution of the issue raised by the present appeals – that prior authorisation of the non-continuation of the treatment should, as a matter of good practice, be obtained in the Family Division of the High Court. In making the latter recommendation the House was reflecting its conclusion In re F (Mental Patient: Sterilisation) [1990] 2 AC 1, at 56 and 79, reached in the light of a review of practice in the U.S. and Australia, that an operation of sterilisation should not be performed on an incapable adult without prior judicial authorisation.

I agree with the observation of Lord Neuberger at para 94 that, in sanctioning a course leading to the death of a person about which he was unable to have a voice, the decision in the Bland case was arguably more extreme than any step which might be taken towards enabling a person of full capacity to exercise what must, at any rate now, in the light of the effect given to article 8 of the ECHR in the Haas case at para 51, cited at para 29 above, be regarded as a positive legal right to commit suicide. Lord Sumption suggests in para 212-213 below that it remains morally wrong and contrary to public policy for a person to commit suicide. Blackstone, in his Commentaries on the Laws of England, Book 4, Chapter 14, wrote that suicide was also a spiritual offence "in evading the prerogative of the Almighty, and rushing into his immediate presence uncalled for". If expressed in modern religious terms, that view would still command substantial support and a moral argument against committing suicide could convincingly be cast in entirely non-religious terms. Whether, however, it can be elevated into an overall conclusion about moral wrong and public policy is much more difficult.

The second objection relates to the so-called "slippery slope". In respectful disagreement with Lord Kerr at para 354 below, I consider that, unless the court can be satisfied that any exception to the subsection can be operated in such a way as to generate an acceptably small risk that assistance will be afforded to those vulnerable to pressure to seek to commit suicide, it cannot conclude that the absolute prohibition in the subsection is disproportionate to its legitimate aim. In this respect the court may already be confident; but it cannot be satisfied. In an area in which the community would expect its unelected judiciary to tread with the utmost caution, it has to be said that, in appeals which the Court of Appeal understood to be presented to it on the basis that Mr Lamb could not commit, and that the late Mr Nicklinson could not have committed, suicide even with assistance, with the result that the issue which it addressed was their alleged right to euthanasia, the evidence and argument available to this court fall short of enabling it to be satisfied of what, like Lord Neuberger, I regard as a pre-requisite of its making a declaration of incompatibility.

Were Parliament for whatever reason, to fail satisfactorily to address the issue whether to amend the subsection to permit assistance to be given to persons in the situation of Mr Nicklinson and Mr Lamb, the issue of a fresh claim for a declaration is to be anticipated. It would no doubt be issued, as was that of Mr Nicklinson, in the Family Division of the High Court. The Crown would be entitled pursuant to section 5(1) of the 1998 Act to notice of the claim and I expect that the Attorney General would thereupon see fit to intervene pursuant to section 5(2). In that way the court would, I hope, receive the focussed evidence and submissions which this court has lacked. While the conclusion of the proceedings can in no way be prejudged, there is a real prospect of their success. Two features of a declaration are worth noting.

The first is that it is indeed legitimate for a declaration to be made even though the provision only sometimes operates incompatibly with human rights. Thus in the Bellinger case, cited by Lord Neuberger at para 114 above, the former provision in section 11(c) of the Matrimonial Causes Act 1973, namely that a marriage shall be void if the parties are not respectively male and female, was declared incompatible even though it infringed the rights under article 8 only of those who had undergone gender reassignment and wished to marry persons of their own genetic sex. The concomitant is, however, that, in making a declaration, it behoves the court precisely to identify in the circumstances of the successful applicant the factors which precipitate the provision's infringement of his human rights. In addressing its task of fashioning a response to the declaration, Parliament deserves no less.

The second, linked, feature of a declaration is that it affords to the courts of the U.K., no doubt uniquely, an opportunity to collaborate to some extent with Parliament in the amendment of the statutory provision which is discovered to have overridden human rights. I do not regard a degree of collaboration as objectionable or, in particular, as compromising judicial independence. But a court will be of maximum assistance to Parliament in this regard if it not only identifies the factors which precipitate the infringement but articulates options for its elimination.

In this latter regard I wish expressly to indorse Lord Neuberger's suggestion at para 123 that, in formulating an exception to the subsection, Parliament might adopt the procedure approved in the F and Bland cases and require that a High Court judge first be satisfied that a person's wish to commit suicide was (to use words which Parliament may feel able to improve) voluntary, clear, settled and informed. I am unaware of any situation in which the courts have acknowledged an inability to distinguish between the expression of an intention which genuinely reflects the speaker's wish and one which does not do so. The ways in which the intentions have been expressed; the consistency or otherwise of its expression; the explanation proffered for it; and, of course, the quality of the speaker's life; all these would inform the court's inquiry. A court might wish to hear evidence from the claimant himself, directly or indirectly; from members of his family; from his friends; from his medical practitioner and other professionals involved in his care; and no doubt also from a doctor and/or psychiatrist and/or other medical expert introduced into the case in order to report to the court. As a former judge of the Family Division, but with hesitation apt to the absence of submissions in this regard, I identify the following factors which the court might wish to investigate before deciding whether it can be so satisfied:

(a) the claimant's capacity to reach a voluntary, clear, settled and informed decision to commit suicide and the existence of any factor which, notwithstanding the requisite capacity, might disable him from reaching such a decision;

(b) the nature of his illness, physical incapacity or other physical condition ("the condition");

(c) the aetiology of the condition;

(d) its history and the nature of the treatments administered for it;

(e) the nature and extent of the care and support with which the condition requires that he be provided;

(f) the nature and extent of the pain, of the suffering both physical and psychological and of the disability, which the condition causes to him and the extent to which they can be alleviated;

(g) his ability to continue to tolerate them and the reasonableness or otherwise of expecting him to continue to do so;

(h) the prognosis for any change in the condition;

(i) his expectation of life;

(j) his reasons for wishing to commit suicide;

(k) the length of time for which he has wished to do so and the consistency of his wish to do so;

(l) the nature and extent of his discussions with others, and of the professional advice given to him, about his proposed suicide and all other options for his future;

(m) the attitude, express or implied, to his proposed suicide on the part of anyone likely to benefit, whether financially or otherwise, from his death;

(n) the proposed mechanism of suicide and his proposed role in achieving it;

(o) the nature of the assistance proposed to be given to him in achieving it;

(p) the identity of the person who proposes to give the assistance and the relationship of such person to him;

(q) the motive of such person in proposing to give the assistance; and

(r) any financial recompense or other benefit likely to be received by such person in return for, or in consequence of, the proposed assistance.

Lord Neuberger comments at para 118 that it may be somewhat premature for me to identify the above factors. But, in that a majority of the court expects that even now, prior to the making of any declaration, Parliament will at least consider reform of the law, I put forward the factors with a view only to enabling Parliament to appreciate the scrupulous nature of any factual inquiry which it might see fit to entrust to the judges of the Division.

On balance I concur in upholding the appeal of the Director of Public Prosecutions in the proceedings brought by Martin and in dismissing his cross-appeal. By issue of the current policy, the director has done all that the House of Lords required in the Purdy case, cited at para 39 above. There is certainly a case for concluding that she might reasonably do more to clarify, in one way or another, the size of the risk that she would consent to the prosecution of health care professionals who, out of a sense of professional concern, perhaps even of perceived obligation and in any event of sympathy, propose to relieve their patients of profound and permanent suffering by assisting them to commit suicide. But big questions are raised, particularly in the judgment of Lord Hughes below, whether the fact that she might reasonably do more can properly be translated by the principle of legality in article 8 into a legal obligation. A more satisfactory outcome for the health care professionals than more detailed exposition of the director's policy would be a court's conclusion that their proposed assistance falls within a statutory exception to the prohibition in the subsection. By the judgments of five members of this court in the other appeals, the prospect of some such exception has come at least somewhat closer and, were it to materialise, it would represent a resolution to the unenviable difficulties currently confronting them which would be sounder in law as well as more satisfactory to themselves.

LORD SUMPTION

Introduction: assisted suicide

English judges tend to avoid addressing the moral foundations of law. It is not their function to lay down principles of morality, and the attempt leads to large generalisations which are commonly thought to be unhelpful. In some cases, however, it is unavoidable. This is one of them.

Suicide is not a novel issue. The moral and legal objections to it have been debated for centuries. There is a case for saying that the only proper concern of the law is to ensure that a person who commits suicide or tries to do so is in a position to make an informed and rational choice. It is the same case today as it was two millennia ago when Seneca described suicide as the last defence of a free man against intolerable suffering:

"It makes a great deal of difference whether a man is lengthening his life or only his death. If the body is useless for service, then why should he not free the struggling soul? Perhaps he should even do it a little before he needs to, lest when the time comes he may be unable to perform the act. Since the danger of living in wretchedness is so much greater than the danger of dying soon, he is a fool who refuses to sacrifice a little time to win so much. Few men have lasted through extreme old age to death without impairment, and many have lain inert and useless. How much more cruel, then, do you suppose it really is to have lost a portion of your life, than to have lost your right to end it?": Ep. LVIII.

This is the classic statement of the principle of autonomy. But it expresses only one side of a complex moral dilemma. There are some moral values, of which the state is the proper guardian, with no rational or utilitarian justification, but which are nevertheless accepted because they are fundamental to our humanity and to our respect for our own kind. The principle of autonomy is one of these values. Its basis is the moral instinct, which is broadly accepted by English law subject to well-defined exceptions, that individuals are entitled to be the masters of their own fate. Others are bound to respect their autonomy because it is an essential part of their dignity as human beings.

There is, however, another fundamental moral value, namely the sanctity of life. A reverence for human life for its own sake is probably the most fundamental of all human social values. It is common to all civilised societies, all developed legal systems and all internationally recognised statements of human rights. I cannot put the point better than Hoffmann LJ did in the Court of Appeal in Airedale NHS Trust v Bland [1993] AC 789, 826C-E:

"we have a strong feeling that there is an intrinsic value in human life, irrespective of whether it is valuable to the person concerned or indeed to anyone else. Those who adhere to religious faiths which believe in the sanctity of all God's creation and in particular that human life was created in the image of God himself will have no difficulty with the concept of the intrinsic value of human life. But even those without any religious belief think in the same way. In a case like this we should not try to analyse the rationality of such feelings. What matters is that, in one form or another, they form part of almost everyone's intuitive values. No law which ignores them can possibly hope to be acceptable."

Leaving aside purely regulatory offences, the criminal law necessarily responds to moral imperatives which command general acceptance among the population at large. The problem in this case is that on the issue of suicide, our most fundamental moral instincts conflict. Our belief in the sanctity of life is not consistent with our belief in the dignity and autonomy of the individual in a case where the individual, being of sound mind and full capacity, has taken a rational decision to kill himself.

These are ancient dilemmas. Ours is not the first generation to confront them. But they are more acute and controversial today, for two main reasons, which are related. One is that advances of medical science have made it possible to preserve life well beyond the point where it is worth living. The other is that it is more difficult in modern conditions for intensely personal end-of-life choices to be made informally, within the family and with the support of a trusted medical practitioner. The medical profession, for wholly understandable reasons, is less willing in a transparent, highly regulated and litigious world to take the responsibility for cutting life short or helping someone else to do so, without an assurance of immunity which in the present state of the law is impossible to give.

The answer which English law gives to these questions is entirely clear. Suicide was a common law offence in England until 1961. It was treated as a form of murder. A particular feature of the law of murder, which makes it unusual among offences against the person, is that the consent of the victim is not a defence to a charge of deliberate killing. Suicide, or "self-murder", was therefore an offence notwithstanding its voluntary character. It followed that an unsuccessful attempt at suicide was criminal, and so was the act of an accessory. The Suicide Act 1961 abolished the rule of law which made suicide an offence, but preserved the criminal liability of accessories. As amended by the Coroners and Justice Act 2009, section 2(1) created a statutory offence committed by any person who does an act which is (a) "capable of encouraging or assisting the suicide or attempted suicide of another person", and (b) "intended to encourage or assist suicide or an attempt at suicide."

The reason for decriminalising suicide was not that suicide had become morally acceptable. It was that imposing criminal sanctions was inhumane and ineffective. It was inhumane because the old law could be enforced only against those who had tried to kill themselves but failed. The idea of taking these desperate and unhappy individuals from their hospital beds and punishing them for the attempt was as morally repugnant as the act of suicide itself. It was ineffective because assuming that they truly intended to die, criminal sanctions were incapable by definition of deterring them. For these reasons, attempted suicide had probably never been an offence in Scotland and by 1961 had long ceased to be one in most European countries. Even in England, prosecution had become rare by the time that the offence was abolished. These points are discussed in Glanville Williams, The Sanctity of Life and the Criminal Law (1958), 248-249. However, the continuing legal objection to suicide was reflected in the fact that very many countries in which suicide was lawful nevertheless imposed criminal liability on those who advised or assisted it. Research summarised in the judgment of the European Court of Human Rights in Koch v Germany (2013) 56 EHRR 6 at para 26 suggests that of the 42 members states of the Council of Europe for which information was available, 36 imposed criminal liability on any form of assistance to suicide and another two, while not imposing criminal liability on direct assistance in suicide, prohibited the prescribing of drugs in order to facilitate it. In Haas v Switzerland (2011) 53 EHRR 33, at para 55 the Court concluded that "the vast majority of member states seem to attach more weight to the protection of the individual's life than to his or her right to terminate it."

In R (Pretty) v Director of Public Prosecutions [2002] 1 AC 800, Lord Bingham said at para 35 that, "while the 1961 Act abrogated the rule of law whereby it was a crime for a person to commit (or attempt to commit) suicide, it conferred no right on anyone to do so". Lord Hope, in the same vein, observed at para 106 that the Act "did not create a right to commit suicide." It followed, as both of them pointed out, from the continuing prohibition of advice and assistance under section 2 of the Act. By this they were plainly not seeking to suggest that suicide remained a legal wrong. The point was that it belonged to the familiar category of acts lawful in themselves but contrary to public policy. This is a categorisation which primarily affects the legal responsibilities of third parties. In particular, it has consequences for the criminal liability of secondary parties or for the enforceability of associated contractual and other legal obligations.

The different legal treatment of the person who wishes to commit suicide and the person who is willing to assist him is not arbitrary. It responds to the same moral instincts which give rise to most dilemmas in this field. Recommendation 1418 (1999) of the Council of Europe recorded at paragraph 9c the Council's view that "a terminally ill or dying person's wish to die never constitutes any legal claim to die at the hand of another person," and that "a terminally ill or dying person's wish to die cannot of itself constitute a legal justification to carry out actions intended to bring about death." This is because, as Lord Hobhouse observed in his speech in Pretty at para 111, the intervention of another party "puts the conduct into a different category from conduct which has involved the deceased alone." I think that Hoffmann LJ came close to the heart of the matter in Airedale NHS Trust, when he pointed out (at page 831) that this was,

"connected with our view that the sanctity of life entails its inviolability by an outsider. Subject to exceptions like self-defence, human life is inviolate even if the person in question has consented to its violation. That is why although suicide is not a crime, assisting someone to commit suicide is."

Why should this be so? There are at least three reasons why the moral position of the suicide (whom I will call "the patient" from this point on, although the term may not always be apt) is different from that of a third party who helps him to kill himself. In the first place, the moral quality of their decisions is different. A desire to die can only result from an overpowering negative impulse arising from perceived incapacity, failure or pain. This is an extreme state which is unlikely to be shared by the third party who assists. Even if the assister is moved by pure compassion, he inevitably has a greater degree of detachment. This must in particular be true of professionals such as doctors, from whom a high degree of professional objectivity is expected, even in situations of great emotional difficulty. Secondly, whatever right a person may have to put an end to his own life depends on the principle of autonomy, which leaves the disposal of his life to him. The right of a third party to assist cannot depend on that principle. It is essentially based on the mitigating effect of his compassionate motive. Yet not everyone seeking to end his life is equally deserving of compassion. The choice made by a person to kill himself is morally the same whether he does it because he is old or terminally ill, or because he is young and healthy but fed up with life. In both cases his desire to commit suicide may be equally justified by his autonomy. But the choice made by a third party who intervenes to help him is very different. The element of compassion is much stronger in the former category than in the latter. Third, the involvement of a third party raises the problem of the effect on other vulnerable people, which the unaided suicide does not. If it is lawful for a third party to encourage or assist the suicide of a person who has chosen death with a clear head, free of external pressures, the potential arises for him to encourage or assist others who are in a less good position to decide. Again, this is a more significant factor in the case of professionals, such as doctors or carers, who encounter these dilemmas regularly, than it is in the case of, say, family members confronting them for what will probably be the only time in their lives.

The Nicklinson and Lamb appeal: Is section 2 of the Suicide Act in principle compatible with the Human Rights Convention?

The sole directly relevant authority is Pretty v United Kingdom (2002) 35 EHRR 1. Mrs Pretty suffered from motor neurone disease. She wanted to be able to count on the assistance of her husband to commit suicide when her suffering became intolerable to her and she was no longer capable of reaching the Dignitas clinic in Switzerland unaided. The European Court of Human Rights held that section 2 of the Suicide Act, by interfering with Mrs Pretty's right to end her life, engaged article 8.1 of the Convention. In its subsequent decision in Haas v Switzerland (2011) 53 EHRR 33, at paras 50-51, the Court held that the effect of this decision was that

"an individual's right to decide by what means and at what point his or her life will end, provided he or she is capable of freely reaching a decision on this question and acting in consequence, is one of the aspects of the right to respect for private life within the meaning of article 8 of the Convention."

Article 8.1 was engaged because respect for Mrs. Pretty's private life entailed accepting her autonomy in making her own end-of-life choices. This is not exactly a right to commit suicide. It is an immunity from interference by the state with the settled decision of a person of full legal and mental capacity to kill himself, unless the interference can be justified under article 8.2.

That being so, the question arose whether the prohibition of all acts of assistance by section 2 of the Suicide Act was justifiable under article 8.2. In that context, the question could not be addressed simply on the footing that her autonomy entitled her to choose death. She needed the assistance of a third party whose own position had to be considered. Of the three considerations that I have summarised above (paragraph 215), it was the third which the Court regarded as decisive. It was held that section 2 of the Suicide Act was justifiable by considerations of public health and in particular by the implications for vulnerable people. The relevant considerations were summarised as follows at para 74:

"[T]he Court finds... that States are entitled to regulate through the operation of the general criminal law activities which are detrimental to the life and safety of other individuals…. The more serious the harm involved the more heavily will weigh in the balance considerations of public health and safety against the countervailing principle of personal autonomy. The law in issue in this case, section 2 of the 1961 Act, was designed to safeguard life by protecting the weak and vulnerable and especially those who are not in a condition to take informed decisions against acts intended to end life or to assist in ending life. Doubtless the condition of terminally ill individuals will vary. But many will be vulnerable and it is the vulnerability of the class which provides the rationale for the law in question. It is primarily for States to assess the risk and the likely incidence of abuse if the general prohibition on assisted suicides were relaxed or if exceptions were to be created. Clear risks of abuse do exist, notwithstanding arguments as to the possibility of safeguards and protective procedures."

After a brief discussion of the question whether this analysis would create a dangerous precedent, the Court concluded, at para 76:

"The Court does not consider therefore that the blanket nature of the ban on assisted suicide is disproportionate."

In my opinion the passages which I have quoted express the ratio of this decision. The question whether to impose a 'blanket' ban on assisted suicide lay within the margin of appreciation of the United Kingdom. This was because it was for each state to "assess the risk and likely incidence of abuse if the general prohibition on assisted suicide were relaxed or if exceptions were to be created." Section 2 was capable of being justified because although it applied to many people who were not in need of protection, it was open to the United Kingdom to take the view that it had to apply generally in order to serve the needs of those who were.

It is clear from the way in which the Court treated the separate complaint of a contravention of article 14 that it considered that the United Kingdom had taken that view and been entitled to do so. At para 89, the Court wrote:

"Even if the principle derived from the Thlimmenos case is applied to the applicant's situation, however, there is, in the Court's view, objective and reasonable justification for not distinguishing in law between those who are and those who are not physically capable of committing suicide. Under article 8 of the Convention, the Court has found that there are sound reasons for not introducing into the law exceptions to cater for those who are deemed not to be vulnerable. Similar cogent reasons exist under article 14 for not seeking to distinguish between those who are able and those who are unable to commit suicide unaided. The borderline between the two categories will often be a very fine one and to seek to build into the law an exemption for those judged to be incapable of committing suicide would seriously undermine the protection of life which the 1961 Act was intended to safeguard and greatly increase the risk of abuse."

The same conclusion had been reached for substantially the same reasons by the Supreme Court of Canada, dealing with a very similar issue in Rodriguez v Attorney-General of Canada [1993] 3 SCR 519, which the Strasbourg Court regarded as persuasive in Pretty: see para 74. Section 7 of the Canadian Charter of Rights and Freedoms provided that every person had "the right to life, liberty and security of the person and the right not to be deprived thereof except in accordance with the principles of fundamental justice." The Court held that the Canadian prohibition of assisted suicide did not violate the Charter. Writing for the majority, Sopinka J held that section 7 was engaged but that it was justified because of the difficulty of protecting the life of others without a blanket ban:

"Given the concerns about abuse that have been expressed and the great difficulty in creating appropriate safeguards to prevent these, it cannot be said that the blanket prohibition on assisted suicide is arbitrary or unfair, or that it is not reflective of fundamental values at play in our society. I am thus unable to find that any principle of fundamental justice is violated by s. 241(b). (p 608)

As I have sought to demonstrate in my discussion of s.7, this protection is grounded on a substantial consensus among western countries, medical organizations and our own Law Reform Commission that in order to effectively protect life and those who are vulnerable in society, a prohibition without exception on the giving of assistance to commit suicide is the best approach. Attempts to fine tune this approach by creating exceptions have been unsatisfactory and have tended to support the theory of the 'slippery slope'. The formulation of safeguards to prevent excesses has been unsatisfactory and has failed to allay fears that a relaxation of the clear standard set by the law will undermine the protection of life and will lead to abuses of the exception." (p 613)

The relevance of prosecutorial discretion

In Pretty, the European Court of Human Rights considered at para 76 the discretionary elements of English criminal proceedings which in practice mitigated the 'blanket' character of the ban on assisted suicide:

"76 ... The Government has stated that flexibility is provided for in individual cases by the fact that consent is needed from the DPP to bring a prosecution and by the fact that a maximum sentence is provided, allowing lesser penalties to be imposed as appropriate. The Select Committee report indicated that between 1981 and 1992 in 22 cases in which 'mercy killing' was an issue, there was only one conviction for murder, with a sentence for life imprisonment, while lesser offences were substituted in the others and most resulted in probation or suspended sentences. It does not appear to be arbitrary to the Court for the law to reflect the importance of the right to life, by prohibiting assisted suicide while providing for a system of enforcement and adjudication which allows due regard to be given in each particular case to the public interest in bringing a prosecution, as well as to the fair and proper requirements of retribution and deterrence.

77 Nor in the circumstances is there anything disproportionate in the refusal of the DPP to give an advance undertaking that no prosecution would be brought against the applicant's husband. Strong arguments based on the rule of law could be raised against any claim by the executive to exempt individuals or classes of individuals from the operation of the law. In any event, the seriousness of the act for which immunity was claimed was such that the decision of the DPP to refuse the undertaking sought in the present case cannot be said to be arbitrary or unreasonable.

78 The Court concludes that the interference in this case may be justified as 'necessary in a democratic society' for the protection of the rights of others and, accordingly, that there has been no violation of article 8 of the Convention."

I do not read these observations as making the conformity of section 2 with article 8 dependent on the existence of a prosecutorial discretion or the way that it is exercised. The conformity of section 2 with article 8 depended, as I have pointed out, on whether the state's assessment of the "risk and likely incidence of abuse" was such as to justify a blanket ban. This is the sole factor identified at paras 74 and 89 of the Court's judgment. The existence and limits of the prosecutorial discretion are put forward at para 76 (i) as matters which a member state may properly take into account in deciding whether a blanket ban on assisted suicide is proportionate, and (ii) as a reason for rejecting Mrs. Pretty's complaint that the Director of Public Prosecutions had refused to give her an advance undertaking not to prosecute her husband if he helped her to kill herself.

Applying the margin of appreciation

It follows that it is for the United Kingdom to decide whether in the light of its own values and conditions section 2 of the Suicide Act is justifiable under article 8.2 of the Convention in the interest of the protection of health. That gives rise to two issues of principle. One is the nature of the decision, and in particular the extent to which the evidence requires the conformity of section 2 with article 8 to be reassessed. The other is whether in a case with the particular features of this one such a reassessment is a proper constitutional function of the Courts as opposed to Parliament.

The role of evidence

The evidence before us of the risk of abuse if the rule against assisted suicide were to be relaxed or qualified consists substantially of material from two sources: the report of Lord Falconer's Commission on Assisted Dying, and the decision of Lynn Smith J in the Supreme Court of British Columbia on a very similar issue in Carter v Canada [2012] BCSC 886. We were invited to conclude on the basis of this material that the views of Parliament in 1961 and of the Strasbourg Court at the time of Pretty had now been overtaken by the more recent knowledge. Lord Mance has reviewed this material and summarised the problems associated with it in terms with which I agree. There are obvious difficulties about reaching a concluded view on untested, incomplete and second-hand material of this kind. The authority of these sources is also diminished by other considerations. The Commission's report, although measured and, as far as one can tell, objective, was inspired by a campaign to change the law. Committed opponents of assisted suicide declined to give evidence before it. Lynn Smith J's review of the extensive evidence before her excluded a substantial body of apparently relevant material as inadmissible and was ultimately set aside by the Court of Appeal on the ground that it was inconsistent with the law laid down by the Supreme Court of Canada in Rodriguez. However, I would in any event reject the submission that the issue has been overtaken by more recent knowledge because I think that this material even if taken at face value is inconclusive both factually and legally.

It is inconclusive factually, for reasons which emerge very clearly from the report of the Commission on Assisted Dying. The only jurisdictions with experience of legalised assisted suicide are certain states of the United States, of which the most important is Oregon, and the Netherlands, Belgium and Switzerland. The data from these sources is contested and acknowledged to be of variable robustness. It is also sensitive to underlying conditions such as standards of education, the existence of long-term relationships between GPs and patients and other social and cultural factors, which are not necessarily replicated in the United Kingdom. Indeed, there may well be significant regional and sociological variations within the United Kingdom. It is plain from the expert evidence reviewed by the Commission that there is a diversity of opinion about the degree of risk involved in relaxing or qualifying the ban on assisted suicide, but not about its existence. The risk exists and no one appears to regard it as insignificant. There is a reputable body of experienced opinion which regards it as high. It includes the British Geriatrics Society, the British Association of Social Work and Action against Elder Abuse. It may fairly be said that their evidence was not empirical but judgmental and anecdotal. But that may be thought to reflect the nature of the issue, which makes it unrealistic to expect decisive empirical evidence either way.

The concept of "abuse" embraces at least two distinct problems. One is that the boundary between assisted suicide and euthanasia is so porous that in practice it may be crossed too often, sometimes even in cases where there was no true consent. The other is the risk that that if assisted suicide were lawful, some people would be too ready to bring an end to their lives under real or perceived pressure from others.

I can deal shortly with the first kind of abuse. It is true that the boundary between assisted suicide and euthanasia is porous. The point is illustrated by the existence of machines for committing suicide, such as Dr Nitschke's, which involve an elaborate process of production and preparation in which everything is done by the assister apart from the final activation of the equipment which he has set up. There seems to me to be no moral and very little functional distinction between suicide by this method and a lethal injection administered by a third party. Nonetheless, I am sceptical of arguments based on this fact, because they assume that assisters, and in particular medical practitioners, would not understand or respect the boundary between voluntary and involuntary choices or between euthanasia and assistance. The papers for this appeal disclose no evidence to support that assumption and a certain amount of evidence to contradict it. I do not doubt that both assisted suicide and euthanasia occur, but they occur in spite of the present state of the law, and would occur in spite of any safeguards that might be included in some alternative state of the law.

The vulnerability to pressure of the old or terminally ill is a more formidable problem. The problem is not that people may decide to kill themselves who are not fully competent mentally. I am prepared to accept that mental competence is capable of objective assessment by health professionals. The real difficulty is that even the mentally competent may have reasons for deciding to kill themselves which reflect either overt pressure upon them by others or their own assumptions about what others may think or expect. The difficulty is particularly acute in the case of what the Commission on Assisted Dying called "indirect social pressure". This refers to the problems arising from the low self-esteem of many old or severely ill and dependent people, combined with the spontaneous and negative perceptions of patients about the views of those around them. The great majority of people contemplating suicide for health-related reasons, are likely to be acutely conscious that their disabilities make them dependent on others. These disabilities may arise from illness or injury, or indeed (a much larger category) from the advancing infirmity of old age. People in this position are vulnerable. They are often afraid that their lives have become a burden to those around them. The fear may be the result of overt pressure, but may equally arise from a spontaneous tendency to place a low value on their own lives and assume that others do so too. Their feelings of uselessness are likely to be accentuated in those who were once highly active and engaged with those around them, for whom the contrast between now and then must be particularly painful. These assumptions may be mistaken but are none the less powerful for that. The legalisation of assisted suicide would be followed by its progressive normalisation, at any rate among the very old or very ill. In a world where suicide was regarded as just another optional end-of-life choice, the pressures which I have described are likely to become more powerful. It is one thing to assess some one's mental ability to form a judgment, but another to discover their true reasons for the decision which they have made and to assess the quality of those reasons. I very much doubt whether it is possible in the generality of cases to distinguish between those who have spontaneously formed the desire to kill themselves and those who have done so in response to real or imagined pressure arising from the impact of their disabilities on other people. There is a good deal of evidence that this problem exists, that it is significant, and that it is aggravated by negative modern attitudes to old age and sickness-related disability. Those who are vulnerable in this sense are not always easy to identify (there seems to be a consensus that the factors that make them vulnerable are variable and personal, and not susceptible to simple categorisation). It may be, as Lord Neuberger suggests, that these problems can be to some extent be alleviated by applying to cases in which patients wish to be assisted in killing themselves a procedure for obtaining the sanction of a court, such as is currently available for the withdrawal of treatment from patients in a persistent vegetative state. But as he acknowledges, there has been no investigation of that possibility in these proceedings. It seems equally possible that a proper investigation of this possibility would show that the intervention of a court would simply interpose an expensive and time-consuming forensic procedure without addressing the fundamental difficulty, namely that the wishes expressed by a patient in the course of legal proceedings may be as much influenced by covert social pressures as the same wishes expressed to health professionals or family members. These are significant issues affecting many people who are not as intelligent, articulate or determined as Diane Pretty or Tony Nicklinson.

They disclose in turn a more fundamental problem. There is a variety of reasons why the resolution of some issue may lie within the margin of appreciation of the state. It may be because the Strasbourg court has recognised that a legitimate diversity of cultural values among member states of the Council of Europe makes a range of possible answers equally consistent with the Convention. Such issues as the prohibition of abortion in Ireland (A v Ireland (2011) 53 EHRR 13) and the presence of crucifixes in Italian classrooms (Lautsi v Italy (2012) 54 EHRR 3) are cases in point. In cases like these, if the Strasbourg court has held the rule or practice of the particular state to be within the state's margin of appreciation, then absent a fundamental shift of cultural values either within the state in question or among the members states of the Council of Europe generally, there is usually little if any scope for a national court in that state to say that the rule or practice in question is contrary to the Convention. Strasbourg has said that it is not. Different considerations arise if the reason why the rule or practice is within a state's margin of appreciation is that the proportionality of some measure or its rational connection with some legitimate objective in itself is sensitive to national conditions which are more effectively assessed by national institutions. The latter exercise calls for an evaluation by national authorities of local needs and conditions: see Buckley v United Kingdom (1996) 23 EHRR 101 at para 75. But these are not rigid or mutually exclusive categories, and one of the problems about the present issue is that it shares some features of both. The question whether the protection of the health of the vulnerable requires a general prohibition on assistance for suicide cannot be a pure question of fact susceptible to decision on evidence alone. Like many issues in the area of human rights, it turns at least partly on a judgment about the relative importance of the different and competing interests at stake. There is no complete solution to the problem of protecting vulnerable people against an over-ready resort to suicide. I doubt whether even a procedure for obtaining judicial sanction would be a complete solution, although with more information than we have at present it might prove to be a partial one. The real question about all of these possibilities is how much risk to the vulnerable we are prepared to accept in this area in order to facilitate suicide by the invulnerable. This is a particularly difficult balance to draw in a case where the competing interests are both protected by the Convention. For this reason, there is an important element of social policy and moral value-judgment involved. The relative importance of the right to commit suicide and the right of the vulnerable to be protected from overt or covert pressure to kill themselves is inevitably sensitive to a state's most fundamental collective moral and social values.

Parliament or the Courts?

The Human Rights Convention represents an obligation of the United Kingdom. In a matter which lies within the margin of appreciation of the United Kingdom, the Convention is not concerned with the constitutional distribution of the relevant decision-making powers. The United Kingdom may make choices within the margin of appreciation allowed to it by the Convention through whichever is its appropriate constitutional organ. That will depend on its own principles of constitutional law. In In Re G (Adoption: Unmarried Couple) [2009] 1 AC 173, the House of Lords accepted that where questions of social policy were within the United Kingdom's margin of appreciation and admitted of more than one rational choice, that choice would ordinarily be a matter for Parliament, but considered that even in the most delicate areas of social policy, this would not always be so. They held that the rule in question, namely the ineligibility of unmarried couples to adopt children, was irrational and unjustifiably discriminatory because it erected a reasonable generalisation (that children were better brought up by married couples) into a universal rule of eligibility preventing unmarried couples from even being considered. It therefore contravened articles 8 and 14 of the Convention: see paras 16-20 (Lord Hoffmann), 53 (Lord Hope), 129-130, 143-144 (Lord Mance). Doubtless, where there is only one rational choice the Courts must make it, but the converse is not true. Where there is more than one rational choice the question may or may not be for Parliament, depending on the nature of the issue. Is it essentially legislative in nature? Does it by its nature require a democratic mandate? The question whether relaxing or qualifying the current absolute prohibition on assisted suicide would involve unacceptable risks to vulnerable people is in my view a classic example of the kind of issue which should be decided by Parliament. There are, I think, three main reasons. The first is that, as I have suggested, the issue involves a choice between two fundamental but mutually inconsistent moral values, upon which there is at present no consensus in our society. Such choices are inherently legislative in nature. The decision cannot fail to be strongly influenced by the decision-makers' personal opinions about the moral case for assisted suicide. This is entirely appropriate if the decision-makers are those who represent the community at large. It is not appropriate for professional judges. The imposition of their personal opinions on matters of this kind would lack all constitutional legitimacy.

Secondly, Parliament has made the relevant choice. It passed the Suicide Act in 1961, and as recently as 2009 amended section 2 without altering the principle. In recent years there have been a number of bills to decriminalise assistance to suicide, at least in part, but none has been passed into law. Lord Joffe introduced two bills on the House of Lords in 2004 and 2005. The 2005 bill went to a second reading in May 2006, but failed at that stage. Lord Falconer moved an amendment to the Coroners and Justice Bill 2009 to permit assistance to a person wishing to travel to a country where assisted suicide is legal. The amendment also failed. The Assisted Dying Bill, sponsored by Lord Falconer, is currently before the House of Lords. In addition to these specific legislative proposals, the issue of assisted suicide has been the subject of high-profile public debate for many years and has been considered on at least three occasions since 2000 by House of Lords Select Committees. Sometimes, Parliamentary inaction amounts to a decision not to act. But this is not even an issue on which Parliament has been inactive. So far, there has simply not been enough Parliamentary support for a change in the law. The reasons why this is so are irrelevant. That is the current position of the representative body in our constitution. As Lord Bingham observed in R (Countryside Alliance) v Attorney-General [2008] AC 719 at para 45, "[t]he democratic process is liable to be subverted if, on a question of moral and political judgment, opponents of the Act achieve through the courts what they could not achieve in Parliament." Cf. Axa v The Lord Advocate [2012] 1 AC 868 at para 49 (Lord Hope).

Third, the Parliamentary process is a better way of resolving issues involving controversial and complex questions of fact arising out of moral and social dilemmas. The legislature has access to a fuller range of expert judgment and experience than forensic litigation can possibly provide. It is better able to take account of the interests of groups not represented or not sufficiently represented before the court in resolving what is surely a classic "polycentric problem". But, perhaps critically in a case like this where firm factual conclusions are elusive, Parliament can legitimately act on an instinctive judgment about what the facts are likely to be in a case where the evidence is inconclusive or slight: see R (Sinclair Collis Ltd) v Secretary of State for Health [2012] QB 394, esp. at para 239 (Lord Neuberger), and Bank Mellat v H.M. Treasury (No. 2) [2013] 3 WLR 179, 222 at paras 93-94 (Lord Reed). Indeed, it can do so in a case where the truth is inherently unknowable, as Lord Bingham thought it was in R (Countryside Alliance) v Attorney-General at para 42.

In the course of argument, it was suggested that the case for the Respondents in the Nicklinson appeal required the Appellants to suffer a painful and degrading death for the sake of others. This is a forensic point, but up to a point it is a legitimate one. It is fair to confront any judge, or indeed legislator, with the moral consequences of his decision. The problem about this submission, however, is that there are many moral consequences of this decision, not all of them pointing in the same direction. For my part, I would accept a less tendentious formulation. In my view, if we were to hold that the pain and degradation likely to be suffered by Mr Lamb and actually suffered by Mr Nicklinson made section 2 of the Suicide Act incompatible with the Convention, then we would have to accept the real possibility that that might give insufficient protection to the generality of vulnerable people approaching the end of their lives. I conclude that those propositions should be rejected, and the question left to the legislature. In my opinion, the legislature could rationally conclude that a blanket ban on assisted suicide was "necessary" in Convention terms, i.e. that it responded to a pressing social need. I express no final view of my own. I merely say that the social and moral dimensions of the issue, its inherent difficulty, and the fact that there is much to be said on both sides make Parliament the proper organ for deciding it. If it were possible to say that Parliament had abdicated the task of addressing the question at all, so that none of the constitutional organs of the state had determined where the United Kingdom stood on the question, other considerations might at least arguably arise. As matters stand, I think it clear that Parliament has determined that for the time being the law should remain as it is.

For this reason I would not wish to encourage the notion that if the case for Mr Nicklinson and Mr Lamb had been differently presented and procedures for scrutinising cases in which patients expressed a desire for assistance in killing themselves had been examined on this appeal, the decision of this court might have been different. In my opinion, the issue is an inherently legislative issue for Parliament, as the representative body in our constitution, to decide. The question what procedures might be available for mitigating the indirect consequences of legalising assisted suicide, what risks such procedures would entail, and whether those risks are acceptable, are not matters which under our constitution a court should decide.

I have not dealt with the possibility that the present state of the law might also be justifiable under article 8.2 for the protection of morals. That is because the point was hardly argued, and because the protection of health seems to me to be a sufficient justification. But I would certainly not rule it out. The criminal law is not a purely utilitarian construct. Offences against the person engage moral considerations which may at least arguably be a sufficient justification for a general statutory prohibition supported by criminal sanctions. The fact that the parties to these proceedings chose not to argue a point which might nevertheless legitimately influence Parliament illustrates one of the difficulties of deciding an issue of this kind judicially in the course of contested forensic litigation.

The Martin appeal: are the Director of Public Prosecution's Guidelines to Prosecutors sufficiently clear?

Although the acts covered by section 2(1) of the Suicide Act constitute an offence in all cases, an important element of discretion is introduced at two stages of the criminal process. The first is the discretion of the Director of Public prosecutions whether to prosecute or consent to a prosecution under section 2(4). The second is the discretion of a sentencing court upon conviction. These discretions are closely related. The Director's decision will be governed by the long-standing practice, published in the Code for Crown Prosecutors and associated guidelines, which requires a prosecutor to be satisfied not only that the evidence is available to justify a conviction, but that it is in the public interest to prosecute. The public interest test depends on the presence of factors mitigating culpability, in other words on the same factors which would be taken into account by a sentencing court if there were a conviction. Indeed the link was once overt. In his classic statement of the policy in 1951, the then Attorney-General Lord Shawcross observed that "it is not always in the public interest to go through the whole process of the criminal law if, at the end of the day, perhaps because of mitigating circumstances, perhaps because of what the defendant has already suffered, only a nominal penalty is likely to be imposed" (Hansard (HC Debates) 483, col 683, 29 January 1951).

I have already expressed the view that section 2 of the Act is compatible with the Convention regardless of the operation of the Director's discretion. There are, however, many circumstances in which the domestic law of a state is not required by the Convention to confer some right or discretion, but nevertheless if it does so, it will be held to the Convention's standards. A Convention state is not required to allow assisted suicide, and if it does, it may qualify it with conditions designed to prevent abuse: Haas v Switzerland (2011) 53 EHRR 33 at paras 57-58. In Gross v Switzerland, (2014) 58 EHRR 7, the European Court of Human Rights held that the ambit of the right and the scope of any restrictions upon it must, within the bounds of practicality, be clear. Therefore in Switzerland, one of the few countries to allow assisted suicide in principle, article 8 was infringed by the Swiss guidelines concerning the circumstances in which medical practitioners might prescribe lethal drugs. This was because they did not sufficiently clearly show how they applied to persons (such as Mrs. Gross) who were not terminally ill.

To be justifiable under article 8.2 of the Convention, a measure engaging article 8.1 must be "in accordance with the law". For this purpose, "law" has an extended definition embracing those respects in which the application of the law depends on practice. In R (Purdy) v Director of Public Prosecutions [2010] 1 AC 345, the House of Lords held that the Code for Crown Prosecutors and any associated guidelines fell within the broad category of "law" for the purpose of deciding whether section 2 of the Suicide Act was justifiable. It followed that the principle of legality required them to be sufficiently accessible and clear, which they were not. It is important to understand what the House regarded as sufficient level of precision and clarity, and why. The problem about law whose application depends on administrative discretion is that, unless the criteria for the exercise of that discretion are made clear in advance, it offers no protection against its inconsistent and arbitrary application. This is the basis of the Strasbourg Court's jurisprudence on the point. As the Court observed in Gülmez v Turkey (Application no 16330/02) (unreported, 20 May 2008), at para 49, "[d]omestic law must afford a measure of protection against arbitrary interference by public authorities with Convention rights, in respect of which the rule of law would not allow unfettered powers to be conferred on the Executive." Lord Hope, with whom Lord Phillips and Lord Neuberger agreed in terms and Baroness Hale and Lord Brown in substance, recognised this in Purdy: see paras 41, 46. He cited as the guiding principle the test stated by the European Court of Rights in Hasan and Chaush v Bulgaria (2002) 34 EHRR 55 at para 84:

"In matters affecting fundamental rights it would be contrary to the rule of law, one of the basic principles of a democratic society enshrined in the Convention, for a legal discretion granted to the executive to be expressed in terms of an unfettered power. Consequently, the law must indicate with sufficient clarity the scope of any such discretion conferred on the competent authorities and the manner of its exercise. The level of precision required of domestic legislation which cannot in any case provide for every eventuality depends to a considerable degree on the content of the instrument in question, the field it is designed to cover and the number and status of those to whom it is addressed."

Lord Hope considered that protection against arbitrary exercises of discretion required that the Director's policy should be stated in advance with sufficient precision to make the consequences of a given course of action reasonably foreseeable.

"The requirement of foreseeability will be satisfied where the person concerned is able to foresee, if need be with appropriate legal advice, the consequences which a given action may entail. A law which confers a discretion is not in itself inconsistent with this requirement, provided the scope of the discretion and the manner of its exercise are indicated with sufficient clarity to give the individual protection against interference which is arbitrary." Para 41

A high standard of clarity and precision is required of any law defining the elements of a criminal offence. We are not, however, concerned with the elements of criminal liability but with the likelihood that those who have incurred criminal liability will be prosecuted. That is not a matter of definition but of discretion. The degree of clarity and precision which it is reasonable to expect of a published policy about the exercise of the prosecutorial discretion is different in at least two important respects from that which can be expected of a statutory provision creating an offence.

The first is that the pursuit of clarity and precision must be kept within the bounds of practicality. What is practically attainable, as the European Court of Human Rights recognised in the passage which Lord Hope quoted from Hasan and Chaush v Bulgaria, (quoted at para 238 above), must depend on the range of people and situations to which it is expected to apply. It is not practically possible for guidelines to prosecutors to give a high level of assurance to persons trying to regulate their conduct if the range of mitigating or aggravating factors, or of combinations of such factors, is too wide and the circumstances affecting the weight to be placed on them too varied for accurate prediction to be possible in advance of the facts.

The second limitation is a point of principle. The pursuit of clarity and precision cannot be allowed to exceed the bounds of constitutional propriety and the rule of law itself. The Code and associated guidelines may be "law" in the expanded sense of the word which is relevant to article 8.2 of the Convention. But they are nevertheless an exercise of executive discretion which cannot be allowed to prevail over the law enacted by Parliament. There is a fine line between, on the one hand, explaining how the discretion is exercised by reference to factors that would tend for or against prosecution; and, on the other hand, writing a charter of exemptions to guide those who are contemplating breaking the law and wish to know how far they can count on impunity in doing so. The more comprehensive and precise the guidelines are, the more likely they are to move from the first thing to the second. As Lord Bingham observed in R (Pretty) v Director of Public Prosecutions [2002] 1 AC 800 at para 39, the Director has no power to give a "proleptic grant of immunity from prosecution". This is not just a limitation on the statutory powers of a particular public official. It is a constitutional limitation arising from the nature of the function which he performs. The Bill of Rights declares that "the pretended Power of Suspending of Laws or the Execution of Laws by Regal Authority without Consent of Parliament is illegal." The European Court of Human Rights expressed the same notion in Pretty at para 77, when it pointed out that "strong arguments based on the rule of law could be raised against any claim by the executive to exempt individuals or classes of individuals from the operation of the law." Mrs Pretty had originally made an extreme claim. She wanted the Director to give her an assurance that her husband would not be prosecuted if he helped her to kill herself. But the point made by the Strasbourg Court would have applied equally, as they pointed out, to a case where the exemption was sought for "classes of individuals", and this must be so whether those classes are defined by their acts or in any other way.

Although both of these limitations emerge clearly from the Strasbourg case-law cited by Lord Hope in support of his analysis in Purdy, neither of them was considered in detail in that case. This was because the published criteria which were held to be inadequate in Purdy were exceptionally vague. They consisted at that stage only in the Code which, because it had to cover the whole range of criminal offences, was necessarily couched in wholly general terms. No one was suggesting that the Director should do more than set out the most significant factors that would guide his decision: see the argument of Lord Pannick QC at page 350B/C. Lord Hope concluded from his examination of the principle that the Director should be required to "to promulgate an offence-specific policy identifying the facts and circumstances which he will take into account in deciding, in a case such as that which Ms Purdy's case exemplifies, whether or not to consent to a prosecution under section 2(1) of the 1961 Act": para 56. Lord Brown of Eaton-under-Heywood considered at para 86 that what was needed was "a custom-built policy statement indicating the various factors for and against prosecution." In the event, the order of the House was made in the precise terms suggested by Lord Hope. Anything more than that would, as it seems to me, have been both impractical and contrary to constitutional principle, both problems of which the Committee was profoundly conscious. The Committee must have regarded the limited form of order which they made as satisfying the principle which they had declared. They must also have appreciated that guidance stating the principles on which the discretion was exercisable and indicating the factors for and against prosecution would not in all cases enable the individual to know in advance whether he would be prosecuted, but only what matters would be taken into account.

The Director's published policy

The Director's current policy is described in her predecessor's Policy for Prosecutors in respect of Cases of Encouraging or Assisting Suicide, published in February 2010 after the decision in Purdy. Lord Neuberger has set out the relevant parts, and I will not do so again. In summary, it lists sixteen "public interest factors tending in favour of prosecution" and six "public interest factors tending against prosecution". The factors tending in favour of prosecution include (6) that "the suspect was not wholly motivated by compassion", (12) "the suspect gave encouragement or assistance to more than one victim who were not known to each other", (13) "the suspect was paid by the victim or those close to the victim for his or her encouragement or assistance", (14) "the suspect was acting in his or her capacity as a medical doctor, nurse, other healthcare professional, a professional carer [whether for payment or not],... and the victim was in his or her care", and (16) "the suspect was acting in his or her capacity as a person involved in the management or as an employee (whether for payment or not) of an organisation or group, a purpose of which is to provide a physical environment (whether for payment or not) in which to allow another to commit suicide". Paragraph 44 recommends a "common sense" approach to the question of personal gain:

"It is possible that the suspect may gain some benefit - financial or otherwise - from the resultant suicide of the victim after his or her act of encouragement or assistance. The critical element is the motive behind the suspect's act. If it is shown that compassion was the only driving force behind his or her actions, the fact that the suspect may have gained some benefit will not usually be treated as a factor tending in favour of prosecution. However, each case must be considered on its own merits and on its own facts."

The factors in favour of and against prosecution are all subject to the general considerations at paragraphs 36-42. For present purposes, it is enough to quote paragraphs 39 and 40:

"39 Assessing the public interest is not simply a matter of adding up the number of factors on each side and seeing which side has the greater number. Each case must be considered on its own facts and on its own merits. Prosecutors must decide the importance of each public interest factor in the circumstances of each case and go on to make an overall assessment. It is quite possible that one factor alone may outweigh a number of other factors which tend in the opposite direction. Although there may be public interest factors tending against prosecution in a particular case, prosecutors should consider whether nonetheless a prosecution should go ahead and for those factors to be put to the court for consideration when sentence is passed.

40 The absence of a factor does not necessarily mean that it should be taken as a factor tending in the opposite direction. For example, just because the victim was not 'under 18 years of age' does not transform the 'factor tending in favour of prosecution' into a 'factor tending against prosecution'."

In formulating the published policy the Director did exactly what the order in Purdy required him to do. In the words of Lord Hope's statement of the principle at para 41, it set out "the scope of the discretion and the manner of its exercise." The Director identified the factors that he would take into account, adding appropriate caveats about the importance of taking each case on its merits and considering the weight to be attached to each factor in the light of all the relevant circumstances. Moreover, the policy was carefully drafted so as avoid the risk of appearing to dispense from the operation of the law in certain cases, by identifying relevant factors rather than categories of persons or acts which would not, or probably not be prosecuted. Unless we are prepared to say that the House of Lords was wrong in Purdy to regard the order which it made as answering the principle which it declared, or unless circumstances have changed in some relevant respect, we should not now say that the February 2010 published policy is inadequate. No relevant change of circumstances has been alleged, and far from regarding the order made in Purdy as wrong, it seems to me to have been soundly based on principle.

Martin's case is that the current guidelines are inadequate because they do not make it sufficiently clear that an assister who has done nothing to encourage the suicide and whose assistance was motivated by nothing but compassion, will not be prosecuted. In particular, he says that they draw an unjustifiable distinction between assistance given by those who are connected to the patient by ties of love and affection (which he calls "Class 1 cases"), and others with no such connection ("Class 2 cases"). Martin accepts that the published policy is sufficiently clear about the former category. Unless there is particular cause for concern, all the factors tend against the prosecution of assisters in this category. They can assume, he says, that they will not be prosecuted. But he says that the position of those without emotional ties to the patient is unclear, especially if they are healthcare professionals or other professional carers. It is to their position that his submissions have been mainly directed.

The Director's published policy has deliberately and rightly not been framed by reference to categories of suspect. But the factors listed do suggest a difference in treatment between those whose assistance is given in a professional capacity, whether as doctors, nurses or carers, and others who are connected by emotional bonds to the patient, in practice generally members of his family. In my view, Martin is wrong to suggest that those in the latter category can count on escaping prosecution. That will depend on all the relevant circumstances, of which the emotional bond may well be the most important but is unlikely to be conclusive. However, the published policy does show that assisters of this kind are less likely to be prosecuted than professionals or other outsiders, other things being equal, which they may not be. Thus, the professional character of an assister's involvement will itself tell in favour of prosecution (factor 14). In addition he, and others without emotional ties to the patient, may not be regarded as "wholly motivated by compassion" (factor 6) and are quite likely to be paid for their assistance (factor 13). Martin objects to these distinctions, as well as to the leeway left to prosecutors by the advice at paragraph 39 that the weight to be given to each factor should be assessed case by case instead of being subject to weightings set out in the published policy. This case was substantially accepted by the majority of the Court of Appeal. But in my view, it was wrong in principle for a number of reasons.

In the first place, although presented as a complaint about the lack of clarity in the published policy, it is in reality a complaint about its substance. As I have pointed out, professionals and other outsiders differ in important respects from those whose willingness to assist the patient arises from an emotional relationship with him. The moral issues raised by the intervention of an outsider are more difficult to assess than those arising from assistance given by (say) members of the patient's family. The answer is likely to be affected by an altogether wider range of factors and therefore to be correspondingly less clear in advance. One can illustrate this by reference to the significance of compassion, which everyone agrees is critical in most of these cases. In the case of a close family member, for example a parent, child or spouse, the compassionate character of his or her motivation will usually be obvious, even if the assister stood to benefit financially by the patient's death. So far as anything is straightforward in this difficult field, it is the overwhelming emotional impact of the patient's suffering on those closest to him. What constitutes a purely compassionate motive in the case of an outsider is likely to be much less obvious. At one extreme, the professional who assists the patient to kill himself may be a long-term living in-carer who has formed an emotional bond with the patient not unlike that of his closest relatives. At the opposite end of the range, the professional may have little or no personal acquaintance with the patient, but out of compassion for human suffering in general holds himself out as being ready to assist patients who have freely chosen suicide. Between these extremes there is an infinitely complex range of possibilities. The position of the professional is likely to be affected by his closeness to the patient, the length of his acquaintance with him, the extent of his previous responsibility for the patient's care, his relations with the patient's family, his opinions about the legal prohibition of assisted suicide, any relevant rules or guidance of his professional body, any involvement on his part in assisting other patients to commit suicide, whether he is paid for his assistance and if so how much, and many other matters. In addition to being more difficult to evaluate, the involvement of the professional raises issues with important implications for other terminally ill or suffering patients, many of whom may be vulnerable. The most that the Director can reasonably be expected to do in the face of such a complex process of evaluative judgment is to identify the main factors that will be relevant. It is neither possible nor proper for him to attempt a precise statement in advance of the facts about when a professional will or will not be prosecuted. Either such a statement will have to be so general and qualified as to be of limited value for predictive purposes, or else it is liable to tie the Director's hands in a way that would in practice amount to a dispensation from the law.

In the Divisional Court Toulson LJ at paras 141-143 gave three reasons why it would be wrong to require the Director to reformulate her policy:

"141. First, it would go beyond the Convention jurisprudence about the meaning of 'law' in the context of the rule of law. Even when considering the meaning of 'law' in the strict sense of that which may be enforced by the courts, the jurisprudence allows a degree of flexibility in the way that it is formulated (Sunday Times v UK). This must apply even more in relation to 'law' in the extended sense of meaning the law as it is liable in practice to be enforced (Purdy paragraph 112), because flexibility is inherent in a discretion. It is enough that the citizen should know the consequences which may well result from a particular course of action.

142. Secondly, it would be impractical, if not impossible, for the DPP to lay down Guidelines which could satisfactorily embrace every person in Mr Havers' class 2, so as to enable that person to be able to tell as a matter of probability whether he or she would be prosecuted in a particular case. As Mr Havers rightly observed, the factors for and against prosecution may point in opposite directions. I do not see how the DPP could be expected to lay down a scheme by which a person would be able to tell in advance in any given case whether a particular factor or combination of factors on one side would be outweighed by a particular factor or a combination of factors on the other side. The DPP is not like an examiner, giving or subtracting marks in order to decide whether a candidate has achieved a pass mark. The DPP has expressed his opposition to any such schematic approach for the good reason that each case ultimately involves a personal judgment.

143. Thirdly, it would require the DPP to cross a constitutional boundary which he should not cross. For the DPP to lay down a scheme by which it could be determined in advance as a matter of probability whether an individual would or would not be prosecuted would be to do that which he had no power to do, i.e. to adopt a policy of non-prosecution in identified classes of case, rather than setting out factors which would guide the exercise of his discretion."

In my opinion, the Court of Appeal had no convincing answer to these points. This is, I think, because there is none.

Ultimately, the question of legal principle posed by the reasoning of the House of Lords in Purdy is whether the uncertainty about the position of professionals allows the arbitrary and inconsistent exercise of executive discretion. In my opinion it does not. Any lack of clarity or precision does not arise from the terms of the Director's published policy. It arises from the discretionary character of the Director's decision, the variety of relevant factors, and the need to vary the weight to be attached to them according to the circumstances of each individual case. All of these are proper and constitutionally necessary features of the system of prosecutorial discretion. The terms of the published policy reflect them. The document sets out the principal relevant factors for and against. It treats the professional character of an assister's involvement as a factor tending in favour of prosecution. It is at least as clear as any sentencing guidelines for this offence could be.

The Lord Chief Justice's interpretation of the Director's published policy

I turn, finally, to a question which arose in the course of the argument, and which has assumed greater prominence than was perhaps expected when the appeal was opened. Lord Judge dissented in the Martin appeal, mainly because he took a different view of the interpretation of the published policy from the rest of the Court of Appeal. Paras 185 and 186 of his judgment have been set out, substantially in full, by Lord Neuberger at para 142 of his judgment. In summary, Lord Judge thought that factor 14 tending in favour of prosecution was concerned only with professionals who abused a position of trust arising from their professional relationship with the patient, for example by bringing undue influence to bear upon him. He thought that it did not extend to a "professional carer who, with no earlier responsibility for the care of the victim, comes in from outside to help." He would have regarded it as an "extraordinary anomaly" that such a person should be more likely to be prosecuted than the family members who brought him in, at any rate if he was not "profiteering". This, in Lord Judge's view, was because such a person would be doing no more than (say) the patient's wife would do if she could. He regarded it as

"an extraordinary anomaly, that those who are brought in to help from outside the family circle, but without the natural love and devotion which obtains within the family circle, are more likely to be prosecuted than a family member when they do no more than replace a loving member of the family, acting out of compassion, who supports the 'victim' to achieve his desired suicide."

Like Lord Neuberger, I do not think that this is what the Director's published policy says. On its face, it discloses a much more general principle that the professional character of an assister's involvement is in all circumstances a factor tending in favour of prosecution, although one whose weight will vary (like all the listed factors) according to the circumstances.

Nonetheless, in the course of argument, Counsel for the Director accepted, on her specific instructions, that paras 185 and 186 of Lord Judge's judgment correctly represented her policy. If this is so, and if, as I consider, the published policy as it stands says something different, then it is clear that the Director is bound to resolve the inconsistency one way or the other. However, I am not prepared to say that she must resolve it by incorporating Lord Judge's interpretation into the published document. I am not prepared to do this for three reasons. First, it is unnecessary. I have no doubt that the Director will in any event wish to review the terms of the published policy in the light of the judgments on this appeal, especially on this point. Secondly, it is legally inappropriate. The Director's duty is to ensure (i) that her published policy is clear, and (ii) that it accurately represents her actual policy. It is not her duty to adopt Lord Judge's interpretation as her policy, and in the absence of her concession in argument nobody could have suggested that it was. Third, it would not be appropriate to make an order the effect of which would be to hold her to that concession, until she has had the fullest opportunity of considering the implications of the two relevant paragraphs of Lord Judge's judgment for the published policy as a whole, in the light of her legal and constitutional role as a prosecuting authority and in the light of our judgments in this case.

The third point requires some expansion. The reason for making it is that Lord Judge's two paragraphs cannot simply be incorporated into the existing published policy. They beg a number of questions, some of them fundamental, on which we do not know the Director's views, and on which she may not yet have formed concluded views. Lord Judge was interpreting factor (14) tending in favour of prosecution. But the relationship between a revised factor (14) and the other factors would need careful consideration if the resulting document is to be clear and coherent. The various listed factors for and against prosecution set out in the existing published policy are concerned with two main matters: (i) whether the assister was entitled to believe that the patient had made a free, settled and unpressured decision to die, and (ii) whether the assister was motivated wholly by compassion. The most difficult issue concerns the relationship between a revised factor (14) and the existing factor (6), which treats it as a factor tending in favour of prosecution that the suspect was "not wholly motivated by compassion". The purely compassionate character of the assister's motivation is a major head of mitigation, which is more likely to be available to someone with an emotional connection to the patient than it is to an outsider with no emotional or even a prior professional connection. There may be very little mitigation available to, say, an assister acting under no compelling pressure arising from a prior relationship with the patient, who has simply been brought in to contribute his technical expertise to the commission of a criminal offence. It can fairly be said that in many cases this approach will deprive those closest to the patient of the means of enabling him to kill himself. This is so. But it is not the object of the published policy to facilitate assisted suicide. Its object is to enable prosecutors to address the main factors which mitigate guilt. Otherwise he is at risk of moving away from the concept of mitigating guilt, and towards that of dispensing certain categories of person from the operation of the Act.

The relationship between a revised factor (14) and the existing factor (6) is probably the most delicate issue, but it is not the only one. What kind of "professional carer with no earlier responsibility for the care of the victim" will be covered by the revised policy? One may infer from the existing factor (12) tending in favour of prosecution that they will not generally include those who held themselves out as giving technical assistance for suicide or who, without holding themselves out, had done it before. Lord Judge appears to have made the (surely realistic) assumption that they would charge for their services, although not in a "profiteering" way. But how would such persons be affected by factor (13), which treats the receipt of payment as a factor tending in favour of prosecution. And what would constitute "profiteering"? Equally delicate questions may arise when one broaches the question what kind of assistance is to be covered by the revised policy. One may infer from factor (16) that the revised policy would not extend to the provision of suicide clinics in the United Kingdom to do what Dignitas does in Switzerland. But highly contentious issues may arise as to the application of the revised policy to some forms of assistance falling well short of that extreme. The context of Lord Judge's remarks and the facts of Martin's case suggest that he was thinking mainly of assistance consisting in accompanying the patient to Dignitas in Switzerland. But the same considerations would not necessarily apply to supplying lethal prescription drugs or specialised equipment. All of these questions might require consultation with the medical professions or even the general public, as occurred before the publication of the current policy.

Unless the Director proposes to modify factors (6), (12)-(14) and (16), the circumstances in which Lord Judge's "professional carer with no earlier responsibility for the care of the victim" will be protected may be far too narrowly confined to justify Lord Judge's prediction at para 186 that they are "most unlikely to be prosecuted." But for present purposes the decisive consideration is that it is a matter for the Director and not for us to decide whether to adopt Lord Judge's interpretation of the policy and if so how and how far to do so.

The present state of the law

The current position may fairly be summarised as follows:

(1) In law, the state is not entitled to intervene to prevent a person of full capacity who has arrived at a settled decision to take his own life from doing so. However, such a person does not have a right to call on a third party to help him to end his life.

(2) A person who is legally and mentally competent is entitled to refuse food and water, and to reject any invasive manipulation of his body or other form of treatment, including artificial feeding, even though without it he will die. If he refuses, medical practitioners must comply with his wishes: Sidaway v. Board of Governors of the Bethlem Royal Hospital and the Maudsley Hospital [1985] AC 871, 904-905; In re F (Mental Patient: Sterilisation) [1990] 2 AC 1; Airedale NHS Trust v Bland [1993] AC 789. A patient (or prospective patient) may express his wishes on these points by an advance decision (or "living will").

(3) A doctor may not advise a patient how to kill himself. But a doctor may give objective advice about the clinical options (such as sedation and other palliative care) which would be available if a patient were to reach a settled decision to kill himself. The doctor is in no danger of incurring criminal liability merely because he agrees in advance to palliate the pain and discomfort involved should the need for it arise. This kind of advice is no more or less than his duty. The law does not countenance assisted suicide, but it does not require medical practitioners to keep a patient in ignorance of the truth lest the truth should encourage him to kill himself. The right to give and receive information is guaranteed by article 10 of the Convention. If the law were not as I have summarised it, I have difficulty in seeing how it could comply.

(4) Medical treatment intended to palliate pain and discomfort is not unlawful only because it has the incidental consequence, however foreseeable, of shortening the patient's life: Airedale NHS Trust v Bland [1993] AC 789, 867D (Lord Goff), 892 (Lord Mustill), R (Pretty) v Director of Public Prosecutions [2002] 1 AC 800, 831H-832A (Lord Steyn).

(5) Whatever may be said about the clarity or lack of it in the Director's published policy, the fact is that prosecutions for encouraging or assisting suicide are rare. Between 1998 and 2011, a total of 215 British citizens appear to have committed suicide with medical assistance at the Dignitas clinic in Switzerland. Not one case has given rise to prosecution. Although cases of assisted suicide or euthanasia are periodically reported to the police (85, we were told, between 1 April 2009 and 1 October 2013) there has been only one recent prosecution for assisting suicide, and that was a particularly serious case.

This state of English law and criminal practice does not of course resolve all of the problems arising from the pain and indignity of the death which was endured by Tony Nicklinson and is now faced by Mr Lamb and Martin. But it is worth reiterating these well-established propositions, because it is clear that many medical professionals are frightened by the law and take an unduly narrow view of what can lawfully be done to relieve the suffering of the terminally ill under the law as it presently stands. Much needless suffering may be occurring as a result. It is right to add that there is a tendency for those who would like to see the existing law changed, to overstate its difficulties. This was particularly evident in the submissions of Dignity and Choice in Dying. It would be unfortunate if this were to narrow yet further the options open to those approaching death, by leading them to believe that the current law and practice is less humane and flexible than it really is.

Conclusion

I would dismiss the appeal of Mrs Nicklinson and Mr Lamb. I would allow the Director's appeal in Martin's case, and dismiss Martin's cross-appeal.

LORD HUGHES

The claimants in these cases, and Mrs Nicklinson's husband before his death, together with some other people in similar positions, see themselves as in a cruel paradox. They have concluded that their lives are not worth living. Whether others in comparable positions would think the same of themselves is, for them, not the point. No one questions their mental capacity to reach the decisions that they have. It is impossible not to understand the depth of their dismay, given the combination of appalling limitation on even the most basic of functions and constant pain. If they were able, unassisted, to commit suicide, they have decided that they would. But their disabilities are so great that they cannot do so, unless they can persuade someone else to help them. Their physical conditions are not likely to be terminal in the near future, so that they will remain unable to achieve their wish for an indefinite period. Whatever the legal position, their appeal for relief will not fail to touch most hearers.

A court can, however, only respond to this appeal by applying the law. That is, of course, not the same as deciding what individual judges would personally like the law to be. Under our constitutional arrangements, firmly entrenched even if largely unwritten, the legislative function is committed to Parliament and courts must not usurp it. Courts do have the necessary function to interpret statutes and to decide what they mean, and to synthesise the different sources of English law, statute law, common law and European. An essential question in this case is whether these latter, properly judicial, functions, can extend to afford the claimants the relief they seek.

In this case there is a perfectly clear Act of Parliament. The Suicide Act 1961 abolished the offence of suicide, so that the suicide himself or herself is no longer committing an offence. But it deliberately, and plainly, created in section 2(1) a separate offence of assisting someone else to commit suicide. True it is that this was fifty years ago, but, even if the law knew a concept of statutory obsolescence, as it does not, this statute was deliberately re-enacted in 2009, after lively public and Parliamentary debate, and after a private member's Bill designed to relax the law had been considered in Parliament and rejected. If anything, the new sections 2A and 2B inserted into the 1961 Act in 2009 somewhat extend the scope of the offence under section 2(1). There is no escape from the fact that unless section 2(1) of the Suicide Act is for some reason or to some extent ineffective, anyone who assists the present claimants or people in like position to commit suicide is guilty of an offence.

The only possible route to qualifying the statute lies in the European Convention on Human Rights ("ECHR"), as part of English law via the Human Rights Act 1998. The argument that it does so in the present cases depends on deploying article 8 of the Convention, by one or other of two possible legal routes. Article 8 provides:

"1. Everyone has the right to respect for his private and family life, his home and his correspondence.

2. There shall be no interference by a public authority with the exercise of this right except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others."

The scheme of article 8 is well known. Like several other articles dealing with so-called qualified rights, it first states an area in which it is concerned to limit State action affecting individuals, and then, by paragraph (2) sets out the qualifications which must be exhibited by State action if it is to be legitimate. Those qualifications are two, the first of legality and the second of justification:

i) the State's action must be 'in accordance with the law'; and

ii) it must be justified as a proportionate means to a legitimate end.

The reach of article 8 can now be seen to be extensive. "Private and family life" undoubtedly covers a wide range of personal activity. There are times when, as a sphere of personal activity is identified as falling within the reach of article 8, it is tempting to say that there is therefore a fundamental right to that particular form of activity. The better view is that the fundamental right is to what article 8.1 actually speaks of - namely respect for private and family life. Whether there is a right to do the particular thing under consideration depends on whether the State is or is not justified in prohibiting it, or placing conditions upon it, and that in turn depends on whether the State's rules meet the requirements of article 8.2. To take a simple example unconnected with the present appeals, the consumption of drugs - whether for reasons of health, pain relief, athletic performance or simple recreation - may well be an aspect of private life within the reach of article 8.1. But it does not follow that there is a fundamental right to take cannabis or steroids, ecstasy or cocaine, still less for others to supply such drugs to would-be users. The great majority of European States prohibit at least some drug usage in the general public interest, and such prohibition is generally more than fully justified under article 8.2.

It is now clear that a person's autonomy in making decisions about how to end his life engages article 8. I agree that it follows that his autonomy in deciding to seek advice or assistance also does so. One ought not, however, in the present cases, to begin with the proposition that an individual has a right to make an end of life decision and to seek assistance in carrying it out. That would be to fall into the error explained in the last paragraph and to assume the answer to these cases. These cases depend not simply on article 8.1 but on its interrelation with article 8.2. And although the claimants in both appeals invoke article 8 they rely on quite different aspects of it.

In the first appeal, of Mrs Nicklinson and Mr Lamb, the issue is now whether the claimant can lawfully engage the assistance of a medical practitioner such as Dr Nitschke to provide a complex machine to deliver a lethal injection, which the claimant can himself activate. Their wish is to undergo this process in England. Their case depends upon the justification limb of article 8.2. They can succeed only if the application of section 2(1) to them in their situations would fail the test of proportionate pursuit of legitimate aim. In effect, they can succeed only if in law the generalised ban upon assisted suicide in section 2(1) is contrary to their article 8 rights.

In the second appeal, AM challenges not section 2(1) but the position of the Director of Public Prosecutions. He contends that the policy statement issued by the Director following the order to provide such which was made in R (Purdy) v Director of Public Prosecutions [2009] UKHL 45, [2010] 1 AC 345 is not enough and must be amplified to deal more specifically with the position of a medical carer who assists, as distinct from a friend or relative who does. He has two distinct arguments. First and principally, he invokes article 8.2 not for the rule of proportionate justification but for the rule of legality. His contention is that a restriction on his private life must, to be in accordance with the law, make it sufficiently foreseeable whether a medical carer will be prosecuted. His secondary contention, advanced on his cross-appeal, is that the Director's policy statement discourages the sort of professional compassionate assistance which he seeks, and that it is thus a disproportionate interference with his article 8 rights. This secondary position involves, like the appeal of Mrs Nicklinson and Mr Lamb, the invocation of the justification limb of article 8.2. This distinction is central to these appeals.

So far as the first appeal is concerned, I have little to add to the reasoning of Lord Sumption, with which I respectfully agree. I also agree with the reasons given in the Court of Appeal, which on this point was unanimous. It is plain that the Strasbourg court has not found that a generalised prohibition on assisting others to commit suicide is a breach of article 8. Given the great preponderance of European States which adopt such a rule, and not least recommendation 1418 of the Council of Europe in 1999, it would have been extremely surprising if it had done so. It is true that Strasbourg thus regards the question as one to be resolved by individual States within their margin of appreciation. But in this country, with our constitutional division of responsibility between Parliament and the courts, this is very clearly a decision which falls to be made by Parliament. For the moment, the balance between the public interest in the protection of the vulnerable and the preservation of life on the one hand and the private interests of those minded to commit suicide on the other has been struck by the 1961 Act, re-enacted in 2009. A change, whether desirable or not, must be for Parliament to make. That is especially so since a change would be likely to call for an infrastructure of safeguards which a court decision could not create.

The position of the DPP

Historically, England and Wales came late to a State public prosecutor - considerably later than Scotland and much later than many European countries. The office of Director of Public Prosecutions was not created until 1879 and then in the face of no little opposition. Leaving aside bodies specially authorised to prosecute in particular areas, such as Local Authorities or Health and Safety Inspectors, for more than a century after this prosecutions remained essentially in the hands of the police, each local force of which was independent of any other. Those local police forces prosecuted either through the office of local solicitors instructed for the purpose, or, later, in some cases through solicitors established by the force for this specific purpose. Although the Director of Public Prosecutions had throughout that time the power to take over a prosecution if he judged it necessary, he had a very limited staff and was concerned only with a small number of the most serious cases. It was only with the Prosecution of Offences Act 1985 that a single body, the Crown Prosecution Service, came into existence with the duty, amongst others, of handling virtually all prosecutions initiated by the police, and the Director became its head.

Whichever has been the body initiating prosecutions, the law of England and Wales has always recognised that a prosecution does not invariably follow acts which in law amount to a criminal offence. A well-known statement of the position is that of the Attorney General, Sir Hartley Shawcross, in 1951, cited by Viscount Dilhorne in the House of Lords in Smedleys Ltd v Breed [1974] AC 839 at 856:

"In 1951 the question was raised whether it was not a basic principle of the rule of law that the operation of the law is automatic where an offence is known or suspected. The then Attorney General, Sir Hartley Shawcross, said: 'It has never been the rule of this country - I hope it never will be - that criminal offences must automatically be the subject of prosecution.' He pointed out that the Attorney General and the Director of Public Prosecutions only intervene to direct a prosecution when they consider it in the public interest to do so and he cited a statement made by Lord Simon in 1925 when he said: ' . . there is no greater nonsense talked about the Attorney General's duty than the suggestion that in all cases the Attorney General ought to decide to prosecute merely because he thinks there is what the lawyers call a case. It is not true and no one who has held the office of Attorney General supposes it is.' Sir Hartley Shawcross's statement was indorsed, I think, by more than one of his successors."

It may be relevant, especially when considering European pronouncements in this area, to note that this general position is not the same in a number of European criminal justice systems. Several of them have embedded either in constitution or criminal code the rule that the Public Prosecutor is under a prima face duty to prosecute when facts amounting to an offence are disclosed. Section 152(2) of the German Code of Criminal Code of Procedure is but one example and article 112 of the Italian constitution another. In Germany, such a domestic rule is regarded as an aspect of the principle of legality. There are broadly similar rules in Austria, Greece, Russia, Poland, Spain, Switzerland and Turkey. Whilst it is certainly true that there are increasingly provisions in many such countries permitting a decision that a prosecution in a particular case is not in the public interest, or authorising diversion to other methods of dealing with proscribed conduct (sometimes described as an aspect of the principle of expediency), this rule remains the default position.

It is essential to identify the scope of the decisions thus being taken by prosecutors in England and Wales over the years. The prosecutor is expected to exercise independent judgment in scrutinising all the myriad facts of each particular case put before him. His power to decide, in the public interest, whether to proceed even where there is a prima facie case that the offence has been committed exists in every class of case, from the most trivial to the most serious. There is nothing in the least unusual in this respect about the offence under section 2(1) of the Suicide Act. The decision to be made is of the same kind as might be made, for example, in the case of a 13 year old caught shoplifting. It might well not be in the public interest to prosecute such a youngster, for example if it was apparent that he had been punished severely by either his parents or his school, or the object stolen had been a twopenny sweet. Conversely, it might be very much in the public interest if there had been longstanding widespread thieving by children in the area, there had been public warnings designed to deter which had failed, or the evidence showed that the child had recruited other younger boys to do the same. Similar decisions may have to be made in relation to offences such as causing death by careless driving. It might be judged not to be in the public interest to prosecute a mother whose careless but comparatively venial mistake at the wheel had resulted in the death of her own child where she was clearly going to bear the guilt for the rest of her life. The case against prosecution might be even stronger if the mother were herself seriously disabled in the same accident. Conversely, prosecution might well be in the public interest if she had been showing off at the wheel, had disregarded warnings to slow down, or she had had previous proven episodes of bad driving.

What is common to all these decisions is that they are made ex post facto and are made individually for the single case under consideration, when all the facts have been investigated and are known. Of course some factors may recur, but in different combinations and of different intensities. Every case is different. Contrast the position when the Director of Public Prosecutions, or any other prosecutor, is asked to state in advance when a particular form of behaviour will result in prosecution and when it will not or may not. Then she is in immediate peril of crossing a constitutional Rubicon. She is in danger of doing one or both of two things. First she is likely to create an advance exemption from the law for a particular group of potential offenders. Second, she is likely in effect to modify the law as laid down in statute or at common law. She has no power to do either of these things. Both are a breach of her constitutional position. She is the head of a branch of the Executive, albeit one with the degree of independence of a non-ministerial government department. As Lord Bingham pointed out in R (Pretty) v Director of Public Prosecutions [2001] UKHL 61 [2002] 1 AC 800 at paragraph 39, the power to dispense with and suspend laws and the execution of law without the consent of Parliament was denied to the Crown and its servants by the Bill of Rights 1689 (1 Will & Mary, sess 2, c 2).

Section 2(4) of the Act, which requires the Director to consent to any prosecution brought under section 2(1) does not begin to alter this position, which is general to all offences, whether her consent is a requirement or not. Provisions requiring that prosecutions be brought only with the consent of the Director, or less frequently with that of the Attorney-General or the Director of Her Majesty's Revenue and Customs, are relatively commonplace. The court was provided with a list of well over 130 statutes containing such stipulations. The number of offences affected is a great deal larger than 130 and they range from river pollution to insider dealing, and from lottery offences to corporate manslaughter. In evidence to the Franks Committee in 1972, the Home Office identified typical overlapping reasons for such provisions to be inserted into statutes: (a) to secure consistency in prosecutions, including where the offence may go wider than the mischief aimed at, (b) to prevent vexatious private prosecutions, (c) to enable account to be taken of mitigating factors, (d) to provide an element of central control in sensitive areas and (e) to enable account to be taken of national security or international considerations. There is no reason to think that section 2(4) was inserted into the Suicide Act with any intention of doing more than keeping the prosecutions in reliable hands. There is no reason at all to suppose that section 2(4) carries with it any greater or different function than the case specific ex post facto judgment described above.

This is the "flexibility", inherent in the requirement for the Director to handle prosecutions for the section 2(1) offence, which the Strasbourg court was considering in Pretty v United Kingdom (2002) 35 EHRR 1. Similarly, the Government there also drew attention to the absence of any mandatory sentence for the offence, thus allowing lesser penalties to be imposed as appropriate. Both the process of abstaining from charging and the process of accepting mitigation in sentence are exercises in flexibility applied after the event to a person who has (or prima facie appears to have) in fact committed the offence, and both are decisions made for the individual case. Flexibility in sentencing was, for the Court, illustrated by the evidence, cited at paragraph 76, that over an eleven or twelve year period most so-called 'mercy killing' cases (charged no doubt usually as manslaughter) had resulted in probation orders or suspended sentences. It was in this context that the Court addressed the article 8.2 rule of legality and went on immediately to say:

"It does not appear to be arbitrary to the Court for the law to reflect the importance of the right to life, by prohibiting assisted suicide while providing for a system of enforcement and adjudication which allows due regard to be given in each particular case to the public interest in bringing a prosecution, as well as to the fair and proper requirements of retribution and deterrence."

There is no occasion to read this observation, as Lord Brown read it in Purdy, as a decision that the generalised prohibition on assisting suicide was only saved from incompatibility with article 8 by the existence of the Director's powers in relation to prosecutions. The juxtaposition of those powers with discretionary sentencing is inconsistent with such a reading. Indeed, given the preponderance of generalised prohibitions on assisted suicide throughout Europe, and without any general prosecutorial discretion still less guidance as to how it might be exercised, it would have been extremely surprising if this had been what the Court was saying.

The seductive argument presented on behalf of the claimant in Purdy contained a vital step which ignored the distinction here set out between examination after the event of all the facts of a case and advance exemption from the law of particular kinds of offending. The case for the claimant was opened in this way, at p 349:

"The discretion conferred on the Director of Public Prosecutions by section 2(4) is integral to the application of the criminal offence created by section 2(1).

…

the flexibility introduced by the consent provisions of section 2(4) was recognised by the European Court of Human Rights in Pretty v United Kingdom (2002) 35 EHRR 1 as an important factor relevant to establishing that the prohibition in section 2(1) was not a disproportionate interference with article 8: see at para 76. Section 2(4), therefore, constitutes parliamentary acknowledgment that there is a category of individuals who, notwithstanding they may have committed the offence under section 2(1), should nevertheless suffer no criminal penalty as a result and whom it is not in the public interest to prosecute."

It is legitimate to say that Parliament no doubt recognised that there might be persons who commit the section 2(1) offence, whom it turns out not to be in the public interest to prosecute. That, however, is true of every offence in the criminal calendar. It is not legitimate to suppose that there is a category of such persons which can be identified in advance by the Director of Public Prosecutions. She cannot do so without crossing the constitutional boundary into either changing the law or giving advance exemption from it to a group of potential offenders.

The basis of the case against the Director both in Purdy and in the present appeal of AM on the legality limb of article 8 is a suggested lack of sufficient foreseeability. The Strasbourg court has made clear that the level of precision which is required of domestic law to meet the principle of legality depends to a considerable degree on the content and that the overriding objective of the principle is to guard against arbitrary executive behaviour: see for example Gillan v United Kingdom [2010] ECHR 28; (2010) 50 EHRR 45. But the foreseeability which any citizen is entitled to expect in relation to the decision of a prosecutor whether or not to institute proceedings is no more but no less than the knowledge that the prosecutor will examine all the facts of any case where an offence has been committed and will decide whether or not it is in the public interest to proceed. No doubt the citizen is entitled to expect more when the question is whether he has committed an offence or not, but in the case of the present appellants it is the settled assumption that they will have done so. The Court of Appeal in the present case appears to have accepted the argument that the legality rule of article 8.2 demanded that a person contemplating assisting someone else to commit suicide should know the answer to the question "What is the likelihood of a prosecution?" (see paragraph 140). But that question cannot be answered without crossing the constitutional boundary between judging each case on its merits according to the public interest and providing something close to an advance exemption in particular circumstances. The Strasbourg court has more than once made clear that the principle of legality does not extend to enabling potential offenders to avoid the application to them of a law which they may wish to avoid: see for example Weber and Saravia v Germany [2006] ECHR 1173; (2008) 46 EHRR SE5 at paragraph 93. In the context of this law, it is the crossing of this constitutional boundary which could properly be described as arbitrary, not the preservation of an individualised ex post facto review of a case.

In Purdy, the House of Lords likewise accepted the legality/foreseeability argument. Lord Hope confined himself to this reasoning, which is the way, so far as I can see, that the case for Mrs Purdy was advanced. Despite some observations which may suggest a view that section 2(1) might in some applications fail to be proportionate (Lady Hale at paras 63-64, Lord Brown at para 74), it is clear that the order made was based on acceptance of the legality/flexibility case (Lady Hale at para 64, first sentence, Lord Brown at para 85, Lord Neuberger at para 106).

Even if, contrary to my respectful view, the order in Purdy was justified, the argument for AM in the present appeal cannot properly be described as anything other than an attempt to obtain for a particular category of persons an advance indication that they will not be prosecuted even though they will have committed the offence. Mr Havers QC rests his case on the contention that the existing policy issued by the Director satisfactorily indicates what may happen to relatives who, out of compassion, assist a patient to commit suicide ("class 1") but does not provide the same indication to professional carers who do so ("class 2"). It is to be noted that this is not really a claim to greater clarity, which might favour either more or fewer prosecutions; rather, it is a claim to a policy of non-prosecution for class 2. The Court of Appeal accepted this argument, finding at paragraph 140 that the existing policy "does not provide medical doctors and other professionals with the kind of steer in class 2 cases that it provides to relatives and close friends acting out of compassion in class 1 cases." But the legitimate functions of the Director of Public Prosecutions do not extend to giving to a particular group of those who, however understandably, are contemplating committing a criminal offence, an advance "steer" as to whether they are likely to be prosecuted, still less an indication that they will not be prosecuted although they have committed the offence.

There are several further difficulties. First, the legitimate prosecutor's function of deciding whether a particular case does or does not warrant proceedings requires a close examination of all its facts. Certainly amongst the relevant facts will be the character and motivation of the potential defendant. But even more important, in most cases, will be to ask what exactly the potential defendant has done. In the context of the section 2(1) offence, an essential factor is the kind of assistance given and what if any degree of encouragement it involved. The argument in Purdy appears to have proceeded on the assumption of only one kind of assistance, namely arranging a journey to a country where assisted suicide is lawful and within it to a respectable clinic where such assistance is provided. But as the facts of the other appeals in the present case show, this is only one of many ways in which the offence under section 2(1) might be committed. What of assistance to travel to Switzerland but in order there to adopt some different method of suicide away from the Dignitas clinic? What of the doctor who prescribes a lethal dose of barbiturates? What of the doctor who does more, and prepares a syringe for his patient to use, or for a relative to use? And what of someone such as Dr Nitschke who assembles for such as Mr Lamb a complicated piece of machinery but himself stops a millimetre or two short of giving the injection. Mr Lamb, no doubt like others, does not wish to travel abroad. He hopes for a dispensation to allow a doctor to assist him in this country. Does the location make any difference? If it does not, would someone who set up a Dignitas-like clinic in an English city be entitled to the same advance steer or not ? Although the argument in Purdy may have centred entirely on a proposed journey to Switzerland, the order made against the Director did not. It required him:

"…to promulgate policy identifying facts and circumstances which he will take into account in deciding whether to consent to prosecution under section 2(1) of the Suicide Act 1961"

When, loyal to that order, the Director set about formulating more detailed policy, which involved a major exercise in public consultation, it will be observed that he eschewed altogether the otherwise central element of the kind of assistance. He was right to do so. It is quite apparent, and appears to be common ground in the present appeal, that to require his successor to give an advance indication of her policy in relation to differing forms of assistance would cross the line into requiring her to re-define the offence, and that that is illegitimate. But simply to pose these questions demonstrates the illegitimacy of the order against her which is sought in the present case. The order made by the Court of Appeal was wholly open-ended, namely a declaration:

"that the…Director of Public Prosecutions (DPP) is in breach of section 6(1) of the Human Rights Act 1998, read with article 8.2 of the European Convention on Human Rights, in that he has made insufficiently foreseeable the consequences, in terms of the exercise of his prosecutorial discretion under section 2(4) of the Suicide Act 1961, of the encouragement or assistance of a suicide or attempted suicide."

Even if this order could be narrowed to limit it to medical or other professional carers, the problem identified above remains, whilst if it were to be thus narrowed it would be shown even more clearly to be directed to advance exemption of particular groups of offenders.

The second difficulty is demonstrated by the first. If it be the law that the Director must provide more specific policy guidance to offer a "steer" to Mr Havers' class 2 professional carers, it is not easy to see why she should not also be required to provide a similar steer to other groups of potential defendants, for example those whose proposed assistance would take one of the possible forms set out in para 281. If for one such group, then it ought to follow for each of the others, and no doubt for many more.

Thirdly, although it can be said that the section 2(1) offence has particular characteristics, it is difficult to see any proper basis, if the Director is required to indicate in advance factors going to prosecution in this case, why the same should not be true of all other criminal offences, in relation to which her function is the same. In fact, the special nature of this offence can be overstated. It is not unique for the law to make it an offence to assist others to do what is not itself a crime, as is demonstrated for example by the offence of living on the earnings of prostitution: prostitution itself, in the absence of public soliciting, is not an offence but living on the earnings of prostitution is, whether or not it involves any element of exploitation. Those who are the 'victims' of crime may in circumstances other than assisting suicide be instigators of it, for example in some cases of forbidden sexual relationships. There may be a number of cases where the victim's article 8 interests are potentially engaged (subject to justification) unless the crime is committed, the recreational user of dangerous drugs who wants a supplier to sell to him may well be an example. It is obvious that there may well be many reluctant offenders in many crimes. But even if it can properly be said that this offence combines features which are not together found elsewhere, it is the fact that the Director controls all but a marginal set of police prosecutions - see section 3(2) of the Prosecution of Offences Act 1985. If it be the law that she can be required to provide a statement of policy as to factors identifying who is likely to be prosecuted in this case, it is difficult to see why the same law does not apply to other offences. Once such a requirement is made, the criminal law is in danger of being diverted from the proper trial process into anticipatory applications for judicial review of the policy, made on hypothetical or uncertain facts by those who seek either to reduce the likelihood of prosecution or to increase it. Such a process subverts the criminal law and encourages satellite litigation.

Like Lord Sumption, I am unable to see that there is any answer to the three reasons given in the Divisional Court by Toulson LJ (paras 141 to 143) why it would be wrong to require the Director to reformulate her policy.

For these reasons, which supplement those of Lord Sumption with which I largely agree, AM cannot properly call for a yet further policy statement from the Director on grounds of legality/foreseeability. For my part, I do not think it is appropriate, for the reasons set out above, for any court to embark upon close construction of the terms of the Director's existing published policy, although if one is to consider it I do not dissent from the analysis set out by Lord Sumption at paragraph 253. I should also record my respectful agreement with those basic propositions of law set out in Lord Sumption's judgment at 255 subparagraphs (1) - (4).

Nor can AM obtain the order which he seeks against the Director on his alternative ground, relying on the justification limb of article 8.2 and on proportionality. This is a repetition of the proportionality argument in the first appeal. If section 2(1) is not disproportionate unless and until Parliament says that it is, then for the same reason the Director cannot be required to "modify" her policy, for that would be to use the Director to change the law. If on the other hand section 2(1) were to be adjudged contrary to article 8 because disproportionate, then the correct remedy would be a declaration of incompatibility; it would still be impermissible, for all the reasons set out above and as explained by Lord Kerr, for the court to use the Director's powers in an individual case to achieve wholesale changes in the law.

I would for these reasons dismiss the appeals of Mrs Nicklinson and Mr Lamb, and the cross appeal of AM, but allow the appeal of the Director of Public Prosecutions.

LORD CLARKE

I agree that, in the first appeal, the appeals of Mrs Nicklinson and Mr Lamb should be dismissed and, in the second appeal, that the DPP's appeal should be allowed and Martin's cross-appeal should be dismissed. I agree that the appeals and cross-appeal should be so disposed of for the reasons given by Lord Sumption, Lord Reed and Lord Hughes. I add a few words of my own on the differing views, not as to the disposal of the appeals, but as to what may happen in the future.

Lord Neuberger, Lord Mance and Lord Wilson conclude that the appeal and cross-appeal should be disposed of in the same way but contemplate the possibility that circumstances may arise in the future in which an application for a declaration of incompatibility might succeed. In his para 197 Lord Wilson has summarised what he calls Lord Neuberger's crucial conclusions in the first appeal. I agree that those are indeed his crucial conclusions. I also agree with the conclusions at para 197(a) to (e). Among the critical factors appear to me to be the fact that the detailed proposals made by Lord Neuberger and Lord Wilson were not advanced in argument and thus have not been subjected to the kind of detailed scrutiny that these difficult questions deserve. A further critical factor is that to date Parliament has not considered the position of those in a similar position to that of Mr Nicklinson and Mr Lamb.

I agree with Lord Wilson that Lord Neuberger also included the points in his para 197(f) and (g). However, he went further, in order to explain what he meant by saying in para 118 (referred to in Lord Wilson's para 197(f)) what might happen if the issue was not 'satisfactorily addressed'. Lord Neuberger said that, for various reasons, one would expect to see the issue whether there should be any and if so what legislation covering those in the situation of the Applicants explicitly debated in the near future. Importantly, he added this:

"Nor would it be possible or appropriate to identify in advance what would constitute satisfactory addressing of the issue, or what would follow once Parliament had debated the issue: that is something which would have to be judged if and when a further application is made. So that there is no misunderstanding, I should add that it may transpire that, even if Parliament did not amend section 2, there should still be no declaration of incompatibility: that is a matter which can only be decided if and when another application is brought for such a declaration. In that connection, Lord Wilson's list of factors in para 205 [above], while of real interest, might fairly be said to be somewhat premature."

Subject to what follows, I agree with Lord Neuberger. If Parliament chooses not to debate these issues, I would expect the court to intervene. If, on the other hand, it does debate them and, after mature consideration, concludes that there should be no change in the law as it stands, as at present advised and save perhaps in exceptional circumstances, I would hold that no declaration of incompatibility should be made. In this regard I agree with the views expressed by Lord Mance at para 190, after referring earlier to the opinion of Rendquist CJ in Washington v Glucksberg 521 US 702 (1997) at p 735, that Parliament is certainly the preferable forum in which any decision should be made, after full investigation and consideration, in a manner which will command popular acceptance. In these circumstances I would conclude that the courts should leave the matter to Parliament to decide. I recognise that it may well be that, for the reasons given by Lord Neuberger and Lord Wilson, Parliament will conclude that some such process as they suggest might be appropriate but, as I see it, that is a matter for it (and not the courts) to determine. In particular, judges should not express their own personal views on the moral questions which arise in deciding what is the best way forward as a matter of policy. As Lord Sumption says in para 228, the imposition of the personal opinions of professional judges in matters of this kind would lack all constitutional legitimacy.

LORD REED

I agree with the majority of the court that, in the first appeal, the appeals of Mrs Nicklinson and Mr Lamb should be dismissed and, in the second appeal, that the DPP's appeal should be allowed and Martin's cross-appeal should be dismissed. In relation to these matters I am generally in agreement with the reasoning of Lord Clarke, Lord Sumption and Lord Hughes, so far as consistent with the following observations of my own. There is also a great deal in the judgment of Lord Mance with which I respectfully agree, including in particular his discussion of proportionality.

I add a few words of my own in order to clarify one mattter. I entirely accept that, as Lady Hale puts it, even if the Strasbourg court would regard the issue before us as within the margin of appreciation which it accords to member states, it is within the jurisdiction accorded to this court under the Human Rights Act 1998 to decide whether the law is or is not compatible with the Convention rights recognised by UK law. If the question whether a provision of primary legislation is compatible with a Convention right arises before one of the courts listed in section 4(5) of the Human Rights Act 1998, the court evidently has jurisdiction to determine it.

In that respect, amongst others, the Human Rights Act introduces a new element into our constitutional law, and entails some adjustment of the respective constitutional roles of the courts, the executive and the legislature. It does not however eliminate the differences between them: differences, for example, in relation to their composition, their expertise, their procedures, their accountability and their legitimacy. Accordingly, it does not alter the fact that certain issues are by their nature more suitable for determination by Government or Parliament than by the courts. In so far as issues of that character are relevant to an assessment of the compatibility of executive action or legislation with Convention rights, that is something which the courts can and do properly take into account. They do so by giving weight to the determination of those issues by the primary decision-maker. There is nothing new about this point. It has often been articulated in the past by referring to a discretionary area of judgment.

The question whether section 2 of the Suicide Act 1961 is incompatible with the Convention turns on whether the interference with article 8 rights is justified on the grounds which have been discussed. That issue raises highly controversial questions of social policy and, in the view of many, moral and religious questions on which there is no consensus. The nature of the issue therefore requires Parliament to be allowed a wide margin of judgment: the considered assessment of an issue of that nature, by an institution which is representative of the citizens of this country and democratically accountable to them, should normally be respected. That is not to say that the courts lack jurisdiction to determine the question: on the contrary, as I have explained. But it means that the courts should attach very considerable weight to Parliament's assessment.

In the present case, I am far from persuaded that that assessment is unjustifiable under the Convention. That is not to say that it is inconceivable that the position could alter in the future: changes in social attitudes, or the evolution of the Convention jurisprudence, could bear on the application of the Convention in this context, as they have done in other contexts in the past. But that is not the position at present.

LADY HALE

There is so much in the comprehensive judgment of Lord Neuberger with which I entirely agree. He has shown that, even if the Strasbourg court would regard the issue before us as within the margin of appreciation which it accords to member states, it is within the jurisdiction accorded to this court under the Human Rights Act 1998 to decide whether the law is or is not compatible with the Convention rights recognised by UK law: Re G (Adoption: Unmarried Couple) [2009] 1 AC 173. Hence both he and Lord Wilson accept that, in the right case and at the right time, it would be open to this court to make a declaration that section 2 of the Suicide Act 1961 is incompatible with the right to respect for private life protected by article 8 of the European Convention on Human Rights. Understandably, however, they would prefer that Parliament have an opportunity of investigating, debating and deciding upon the issue before a court decides whether or not to make such a declaration. Lord Mance is also prepared to contemplate that possibility, although he too thinks Parliament the preferable forum in which any decision should be made (paras 190-191)). Together with Lord Kerr and I, who would make a declaration now, this constitutes a majority who consider that the court both can and should do this in an appropriate case. Lord Clarke (para 293) and Lord Sumption (para 233) might intervene but only if Parliament chooses not to debate the issue; otherwise, they, and Lord Reed and Lord Hughes, consider that this is a matter for Parliament alone.

Like everyone else, I consider that Parliament is much the preferable forum in which the issue should be decided. Indeed, under our constitutional arrangements, it is the only forum in which a solution can be found which will render our law compatible with the Convention rights. None of us consider that section 2 can be read and given effect, under section 3(1) of the Human Rights Act 1998, in such a way as to remove any incompatibility with the rights of those who seek the assistance of others in order to commit suicide. However, in common with Lord Kerr, I have reached the firm conclusion that our law is not compatible with the Convention rights. Having reached that conclusion, I see little to be gained, and much to be lost, by refraining from making a declaration of incompatibility. Parliament is then free to cure that incompatibility, either by a remedial order under section 10 of the Act or (more probably in a case of this importance and sensitivity) by Act of Parliament, or to do nothing. It may do nothing, either because it does not share our view that the present law is incompatible, or because, as a sovereign Parliament, it considers an incompatible law preferable to any alternative.

Why then is the present law incompatible? Not because it contains a general prohibition on assisting or encouraging suicide, but because it fails to admit of any exceptions. The problem with the present law is vividly illustrated by comparing the situation of people like Mr Nicklinson, Mr Lamb and Martin with that of Ms B: see Re B (Consent to Treatment: Capacity) [2002] EWHC 429 (Fam), [2002] 1 FLR 1090.

Ms B was a professional woman in her forties, who became paralysed from the neck down as a result of a cervical cavernoma. She could move her head and use some of her neck muscles but could not move her torso, arms and legs at all. She was totally dependent upon her carers in the intensive care unit where she had been for a year. Her life was supported by artificial ventilation. Without it she would have a less than 1% chance of independent ventilation. And death would almost certainly follow. She wanted the ventilator turned off but her doctors refused to do so. She brought proceedings in the Family Division of the High Court seeking declarations that she had the mental capacity to choose whether or not to accept the treatment and that the hospital was treating her unlawfully, together with nominal damages to recognise the tort of trespass to her person. Dame Elizabeth Butler-Sloss P granted her the remedies she sought. The principal question was whether she had capacity to consent to or refuse life sustaining treatment. If she had that capacity it was for her to make that decision for herself and not for her doctors to make it for her. It was irrelevant whether they or anyone else thought that continued treatment would be in her best interests.

It is important to note that Ms B was entitled to refuse treatment without having to go to court. The hospital should have acceded to her wishes. The only valid reason for not doing so would be a reasonable doubt about whether she had the capacity to give or refuse her consent to life-sustaining treatment. Had she lacked that capacity, the question would indeed have been governed by what was in her best interests. As she did have capacity, she was entitled to take whatever decision she wanted: it was for her to decide where her own best interests lay.

The reason that she had to go to court was that her request for the machine to be turned off was seen by some of the people looking after her as killing her or assisting her to die and thus ethically unacceptable (para 97). But our law draws two crucial distinctions. The most important is between the positive and the negative, between killing and letting die, between taking active steps to end a patient's life, even though this is what the patient herself earnestly desires, and withholding or withdrawing life-sustaining medical treatment or intervention to which a patient refuses her consent (whether at the time or in advance). While this distinction may make sense to us, it must often make little sense, especially to those who suffer the cruel fate of paralysis: those who can breathe without artificial help are denied a choice which those who cannot do so may make, should they wish to do so. For some of the people looking after them, it will be a mystery why they must switch off the machine or withdraw artificial nutrition and hydration if this is what the patient wants, but they may not painlessly administer a lethal dose of medication which the patient wants just as much.

The second distinction is between killing and helping someone to kill herself, between murder (or voluntary manslaughter) and assisting suicide. Both are crimes, but the latter is less serious than the former. The distinction between them is less clear cut than the distinction between killing and letting die, but it is nevertheless important. Mercy killing is the choice and the act of the person who kills, however benevolent the motive. Committing suicide is the choice and the act of the person who does it, and that person commits no crime. Hence, as Lord Neuberger explains, assisting suicide is a very unusual offence.

In Pretty v United Kingdom (2002) 35 EHRR 1, disagreeing with the majority of the House of Lords in R (Pretty) v Director of Public Prosecutions [2001] UKHL 61, [2002] 1 AC 800, the Strasbourg court held that the right to respect for private life protected by article 8.1 of the European Convention on Human Rights was "engaged" by the prohibition of assisting suicide contained in section 2(1) of the Suicide Act 1961 (most clearly stated in para 86, referring back to paras 61 to 67). The court agreed with Lord Hope that "the way she chooses to pass the closing moments of her life is part of the act of living, and she has the right to ask that this too must be respected" (para 64). Since then, the Strasbourg court has been even clearer about what the right entails, in Haas v Switzerland (2011) 53 EHRR 33, at para 51 (repeated in Koch v Germany (2013) 56 EHRR 6, para 52, and Gross v Switzerland (2014) 58 EHRR 7, para 59):

". . . . an individual's right to decide by what means and at what point his or her life will end, provided he or she is capable of freely reaching a decision on this question and acting in consequence, is one of the aspects of the right to respect for private life within the meaning of article 8 of the Convention."

I agree with Lord Kerr that the court was not saying that the right to choose the manner and timing of one's death depends upon being physically capable of carrying out that choice without any assistance. Of course, it does not follow from a person's right to respect for her autonomous choices about how and when she wishes to die that she also has the right to demand to be provided with help from other people. It does not follow from the right to marry and found a family in article 12 of the Convention that a person has a right to be provided with a marriage partner. But it does follow from that right that the state's right to place obstacles in the way of a person who does wish to become a marriage partner is severely limited. In Pretty, Haas, Koch and Gross, the Strasbourg court might have drawn a clear distinction between taking one's own life and having the help of another to do so. The court might have said that, while interfering with a person's right to take her own life would require justification under article 8.2, interfering with that person's freedom to receive the willing help of another in doing so did not require justification. But the court said no such thing. It went on in each case to consider the justifications advanced for interfering with the help which others might wish to give. And in the Gross case, it held that the interference was not justified.

The House of Lords must have taken the same view in R (Purdy) v Director of Public Prosecutions [2009] UKHL 45, [2010] 1 AC 345, when it unanimously accepted that the prohibition of assisting suicide in section 2(1) of the Suicide Act 1961 was an interference with the article 8.1 rights of the would-be suicide. Had it not been such an interference, there would have been no need to look for justification under article 8.2, and the requirement that the interference be "in accordance with the law" would not have arisen. This Court has not been invited to hold that Purdy was wrongly decided and I for one would not be prepared to do so.

It must also follow that no distinction can be drawn between those who could do it all for themselves, but merely prefer to have some help, and those who cannot do it all for themselves. I agree entirely with Lord Kerr (at para 332 of his judgment) that that cannot have been what the Strasbourg Court meant by the reference to being "capable of . . . acting in consequence" of their freely reached decision. The action could include authorising others to act as well as taking action oneself.

The question, therefore, remains as it has always been. Is an outright prohibition of such help a proportionate interference with the right of the individual to choose the manner and timing of her death? As is well known, to be justified, such interference has to be (i) for a legitimate aim which is important enough to justify interfering with a fundamental right, (ii) rationally connected to achieving that aim, (iii) no more than reasonably necessary to achieve it, and (iv) in the light of this, striking a fair balance between the rights of the individual and the interests of the community (see R (Quila) v Secretary of State for the Home Department [2012] 1 AC 621, para 45; Bank Mellat v HM Treasury (No 2) [2013] UKSC 39, [2013] 3 WLR 179, 222, para 20).

The only legitimate aim which has been advanced for this interference is the protection of vulnerable people, those who feel that their lives are worthless or that they are a burden to others and therefore that they ought to end their own lives even though they do not really want to. In terms of article 8.2, this could be put either as the "protection of health" or as the "protection of the rights of others", the right in question being the most important right of all, the right to life protected by article 2. As Lord Sumption points out, an alternative aim might be advanced, as the "protection of morals". Respect for the intrinsic value of all human life is probably the most important principle in Judaeo-Christian morality. It would surely justify an absolute refusal to oblige any person to help another commit suicide. It would not so obviously justify prohibiting those who freely judged that, in the circumstances of a particular case, there was no moral impediment to their assisting suicide. Respect for individual autonomy and human dignity are also important moral principles. The very complexity of the moral argument, amply demonstrated in the material before this court, tells against relying upon this as the legitimate aim of the legislation.

Is it then reasonably necessary to prohibit helping everyone who might want to end their own lives in order to protect those whom we regard as vulnerable to undue pressures to do so? I can understand the argument that it is: how does a person judge which pressures are undue and which are not? We can all understand why people placed in the situation of Mr Nicklinson, Mr Lamb, Martin or Ms B might wish an end to their suffering. But (as I ventured to point out in Purdy, at para 66) there are many other reasons why a person might consider it a sensible and reasonable thing to do. On what basis is it possible to distinguish some of those pressures from others?

That problem is certainly enough to justify a general ban on assisting suicide. But it is difficult to accept that it is sufficient to justify a universal ban, a ban which forces people like Mr Nicklinson, Mr Lamb and Martin to stay alive, not for the sake of protecting themselves, but for the sake of protecting other people. In Pretty, the Strasbourg court rejected the argument that Mrs Pretty was suffering inhuman and degrading treatment contrary to article 3. But no-one who has read the appellants' accounts of their lives and their feelings can doubt that they experience the law's insistence that they stay alive for the sake of others as a form of cruelty.

It would not be beyond the wit of a legal system to devise a process for identifying those people, those few people, who should be allowed help to end their own lives. There would be four essential requirements. They would firstly have to have the capacity to make the decision for themselves. They would secondly have to have reached the decision freely without undue influence from any quarter. They would thirdly have had to reach it with full knowledge of their situation, the options available to them, and the consequences of their decision: that is not the same, as Dame Elizabeth pointed out in Re B (Treatment), as having first-hand experience of those options. And they would fourthly have to be unable, because of physical incapacity or frailty, to put that decision into effect without some help from others. I do not pretend that such cases would always be easy to decide, but the nature of the judgments involved would be no more difficult than those regularly required in the Court of Protection or the Family Division when cases such as Aintree University Hospitals NHS Trust v James [2013] 3 WLR 1299 or Re B (Treatment) come before them.

I mention those courts as the decision-makers, because they are accustomed to dealing with such sensitive life and death questions, some of them (as Lord Neuberger points out) even more dramatic than this. But other bodies, sufficiently neutral and independent of anyone involved with the applicant, and skilled at assessing evidence and competing arguments, could be envisaged. The task would differ from that of the Court of Protection when making decisions on behalf of people who lack capacity, in that there would be no discretion or assessment of the applicant's best interests involved. The whole purpose of the procedure is to respect the autonomous choice of a person who has the capacity to make it. In that respect the task would be very similar to that of Dame Elizabeth Butler-Sloss in Re B (Treatment).

Were there to be such a procedure, it would appear to me to be more than sufficient to protect those vulnerable people whom the present universal prohibition is designed to protect. They simply would not meet the qualifications to be allowed help. The process would not be invoked and even if it were it would not succeed in securing them that help. It would be a more suitably targeted solution than any prosecution policy, however enlightened and humane, could ever be. It would have the merit of resolving the issue in advance rather than relying on ex post facto executive discretion to solve the problem (although it should not preclude the exercise of prosecutorial discretion in a case where prior authorisation had not been obtained).

To the extent that the current universal prohibition prevents those who would qualify under such a procedure from securing the help they need, I consider that it is a disproportionate interference with their right to choose the time and manner of their deaths. It goes much further than is necessary to fulfil its stated aim of protecting the vulnerable. It fails to strike a fair balance between the rights of those who have freely chosen to commit suicide but are unable to do so without some assistance and the interests of the community as a whole.

I understand that Lord Neuberger and Lord Wilson are receptive to that view in principle, but consider that this is not the right occasion or the right time to make a declaration of incompatibility. That is an entirely understandable view, given in particular the original focus of the cases of Mr Nicklinson and Mr Lamb on voluntary euthanasia rather than assisted suicide (as explained in full by Lord Mance). The sort of process which I have suggested above was scarcely touched upon, let alone explored, in evidence or argument. However, the question for us is one of principle rather than fact: once the principle is established, the question for the judge or other tribunal which is asked to authorise the assistance would be one of fact. He or she would have to be satisfied on the evidence that the applicant had freely reached a fully informed decision which she had the capacity to reach and needed the defined help which was available to enable her to put that decision into effect. It is at that point that the evidence relating, for example, to Dr Nitschke's machine, would become relevant and important.

I also understand that Lord Mance would not rule out such a solution, but he considers that we lack the evidence, in particular about the risks to people who need the protection of this law, to justify departing from the view taken by the House of Lords in Pretty. It is worth remembering that the House took the view that article 8 was not engaged at all, and so the observations made about the justification for any interference were strictly obiter dicta. Furthermore, the assertions made about the need to protect vulnerable people were just that: they were no more based on solid evidence than were the assertions to the contrary made, for example, in Carter v Canada [2012] BCSC 886. Indeed, the experience of those few jurisdictions where assisted suicide is permitted provides some means of testing the case for a universal ban.

In my view, the question is one of principle rather than evidence, and in principle it is the interference which requires justification rather than the limited exception which is suggested. The Carter case will be coming before the Supreme Court of Canada, probably later this year, and it will be interesting to see how they approach the issue.

Left to myself, therefore, I would have allowed the first appeal and made a declaration that section 2(1) of the Suicide Act 1961 is incompatible with article 8, to the extent that it does not provide for any exception for people who have made a capacitous, free and fully informed decision to commit suicide but require help to do so. It seems to me that as a general rule, the prohibition is justified. It is the lack of any exception to meet the particular circumstances of the sorts of case before us that is incompatible. I agree with Lord Wilson (para 203) that it is legitimate to make a declaration even though a provision only sometimes operates incompatibly with the convention rights (as in Bellinger v Bellinger [2003] UKHL 21, [2003] 2 AC 467). I am, however, a little bit nervous about his list of "factors" (para 205), because factors are more readily associated with the exercise of a discretion, rather than an issue of fact, which I believe this to be, and some of them are a little suggestive of a "best interests" jurisdiction. But they are helpful in illustrating some of the factual matters which a decision-maker might wish to explore in addressing the four essential requirements which I have outlined at para 314 above.

Turning to the second appeal, the Director of Public Prosecutions is required by the order made in Purdy to clarify what facts and circumstances she will take into account in deciding whether a prosecution is in the public interest. I entirely agree with Lord Neuberger that she should reconsider her policy in the light of the difference of opinion as to its meaning which emerges from the judgments in the Court of Appeal. We were told on her behalf that the Lord Judge CJ's interpretation of her policy was correct. If so, that should be made clear in the policy. People should be able to go to that policy, and not to the judgments in this court, in order to understand it.

Left to myself, I would go further. It seems to me, as it seemed in Purdy, that the policy has two purposes. The first, and uncontroversial, purpose is to make the way in which decisions to prosecute will be taken sufficiently clear to meet the Convention requirement that the interference be "in accordance with the law". This entails accessibility (hence the need to clarify the policy) and foreseeability, as well as consistency and lack of arbitrariness. We can debate endlessly what the Strasbourg court meant, at para 76 of Pretty (quoted by Lord Neuberger at para 32 above) by first stating that the "Court does not consider therefore that the blanket nature of the ban on assisted suicide is disproportionate" and going on to discuss the flexibility of enforcement in the next sentence. It might have been reverting to the "non-arbitrary" requirement of legality. Or it might have been continuing its discussion of proportionality. I ventured to suggest in Purdy (paras 63 and 64) that the policy may have a part to play in securing that section 2(1) does not operate as a disproportionate interference with the right protected by article 8 and now so clearly articulated in Haas v Swizerland. The underlying theme of the factors which the DPP considers relevant to whether a prosecution will be in the public interest is clearly to identify the sort of cases which might be covered by the exception proposed above. The time may therefore be ripe for a review to see whether further progress can be made in that direction without offending against the constitutional prohibition of "dispensing with the laws". But I agree that there is no need to make an order requiring the DPP to conduct a review. She will no doubt be considering the position in the light of the judgments in this Court and in the Court of Appeal.

Hence, I would have allowed the appeal of Mrs Nicklinson and Mr Lamb and made the declaration of incompatibility outlined above. I am content to allow the Director's appeal and to dismiss the cross-appeal in the case of Martin. I also wish to record my agreement with the important statements in para 255(2), (3), and (4) of Lord Sumption's judgment. I have, however, reservations about both the statements in paragraph 255(1), which may require some qualification or elaboration, especially in the light of Savage v South Essex Partnership NHS Foundation Trust [2008] UKHL 74, [2009] 1 AC 681, and Rabone v Pennine Care NHS Foundation Trust [2012] UKSC 2, [2012] 2 AC 72. A policeman is surely entitled to prevent a would-be suicide from jumping off Westminster Bridge.

I should perhaps add that my conclusion is not a question of "imposing the personal opinions of professional judges". As already explained, we have no jurisdiction to impose anything: that is a matter for Parliament alone. We do have jurisdiction, and in some circumstances an obligation, to form a professional opinion, as judges, as to the content of the Convention rights and the compatibility of the present law with them. Our personal opinions, as human beings, on the morality of suicide do not come into it.

LORD KERR

I agree with Lord Neuberger, Lady Hale, Lord Mance, and Lord Wilson that this court has the constitutional authority to issue a declaration of incompatibility. In agreement with Lady Hale, I consider that there is no reason that we should refrain from doing so.

The first appeal

The overarching issue on the first appeal is whether section 2(1) of the Suicide Act 1961 is incompatible with the appellants' rights under article 8 of the European Convention on Human Rights and Fundamental Freedoms (ECHR). If it is incompatible, then it is the duty of this court to say so. That is a duty with which we have been charged by Parliament. And it is a duty from which we cannot be excused by considerations such as that the Director of Public Prosecutions can choose to implement the law in a way that will not infringe the appellants' rights, or that Parliament has debated the issue and has decided not to repeal it. In making that declaration we do not usurp the role of Parliament. On the contrary, we do no more than what Parliament has required us to do.

Scope of the right

In Haas v Switzerland, (2011) 53 EHRR 33 at para 51 the European Court of Human Rights (ECtHR) said:

"… the Court considers that an individual's right to decide by what means and at what point his or her life will end, provided he or she is capable of freely reaching a decision on this question and acting in consequence, is one of the aspects of the right to respect for private life within the meaning of article 8 of the Convention."

This right against unjustified interference with the freedom 'to decide by what means and at what point his or her life will end' does not impose a positive duty on the state. For it to amount to a positive duty there would have to be some claim that the state was required to furnish the assistance, rather than merely tolerate it. There is no question of the appellants claiming that they should be assisted by the state to do what they want to do.

Affirming statements to like effect appear in para 52 of Koch v Germany (2013) 56 EHRR 6 and Gross v Switzerland (2014) 58 EHRR 7, paras 59 and 60.

Nor does this right, contrary to what Lord Sumption suggests in paragraph 215 of his judgment, create a right for a third party to assist. The mere fact that giving effect to the right of the person wishing to receive assistance to die has as a corollary that the assister would not be prosecuted does not mean that the assister has a Convention right to so assist. If that were so, the assister would be able to claim independently that he was entitled to render such assistance. No-one contemplates that.

It is suggested that the words 'capable of … acting in consequence' were carefully devised to exclude from the ambit of article 8 those who are physically incapable of bringing about their desired death. I reject that suggestion. Had it been the Strasbourg court's intention to shut out from the application of article 8 those who wished to end their lives but were physically incapable of doing so, one would surely have expected to have that position explicitly stated and, more importantly, the reasons for it expressly articulated. If some mechanical means (which they could activate) of carrying out their wish was available, they would be capable of 'acting in consequence' of their decision. It cannot seriously be suggested that they are incapable because no such mechanical means exists but that there is available to them willing and informed human intervention.

The only sensible interpretation of this proviso, and the one that accords with common sense, is simply that the person should be capable of exercising free will at all stages of the process. 'Reaching a decision' and 'acting in consequence' are to be read as amounting to this, the emphasis being on 'freely' rather than on a stepwise reading of what it is that one ought freely to be able to do.

Being freely capable of acting on a decision to end one's life does not therefore mean being physically capable of so acting unaided. A person is just as capable of 'freely acting' in consequence of his decision to end his life by recourse to informed and willing assistance to bring that about as he is by drawing exclusively on his own resources. If I wish to die and am physically unable to bring the medication that will end my life to my own lips but have someone who will do that for me, I am acting just as freely by having them do so as if the hand that bore the draught was my own.

The starting point, therefore, is that the appellants have a right under article 8 of ECHR to end their lives and to have recourse to willing, informed assistance to bring about their wish.

The test to be applied

The essential question is therefore whether the interference with that right is justified. Justification of interference with a right to bring intolerable suffering to an end must be of a different order from that which will be required to warrant intervention in most species of article 8 rights. One should not fail to confront the stark reality of this. The appellants are condemned to a life bereft of pleasure or quality. They live in the knowledge of the distress that their condition and their own misery causes to those close to them.

The nature of the interference in this case is not in dispute, and the test for whether it is justified is set out in the decisions of the House of Lords in Huang v Secretary of State for the Home Department [2007] 2 AC 167 and of this court in R (Aguilar Quila) v Secretary of State for the Home Department [2012] 1 AC 621. In the latter case, Lord Wilson said at para 45:

"… In Huang v Secretary of State for the Home Department [2007] 2 AC 167, Lord Bingham of Cornhill suggested, at para 19, that in such a context four questions generally arise, namely: (a) is the legislative objective sufficiently important to justify limiting a fundamental right?; (b) are the measures which have been designed to meet it rationally connected to it?; (c) are they no more than are necessary to accomplish it?; and (d) do they strike a fair balance between the rights of the individual and the interests of the community? …"

Before dealing with the substantive application of the test, however, it is necessary to deal with these preliminary questions about how this court should approach the task: (1) the constitutional relationship between the court and Parliament and (2) the standard of review.

Margin of appreciation and the division of powers in the British constitution

ECtHR's decision in Pretty was that the blanket ban on assisted suicide did not breach Mrs Pretty's rights under article 8 of the Convention. But that does not mean that it was found to be proportionate. As Lord Sumption has said in para 218 of his judgment, the ban was 'capable of being justified because although it applied to many people who were not in need of protection, it was open to the United Kingdom to take the view that it had to apply generally in order to serve the needs of those who were.' The fact that it was capable of being justified and that it was open to the United Kingdom to take the view that the provision had to apply generally was sufficient to withstand Strasbourg's scrutiny because their examination is carried out at one remove from that which this court must apply.

The context in which justification is to be judged is different in the domestic setting. In R (G) (Adoption: Unmarried Couple) [2009] 1 AC 173 it was held that that a fixed rule which excluded unmarried couples from the process of being assessed as potential adoptive parents interfered with their article 8 and article 14 rights. In so finding, the House of Lords said that it should not be inhibited from going further than the European court had gone because a margin of appreciation was available to member states particularly in delicate areas of social policy. At para 32 Lord Hoffmann said:

"It must be remembered that the Strasbourg court is an international court, deciding whether a member state, as a state, has complied with its duty in international law to secure to everyone within its jurisdiction the rights and freedoms guaranteed by the Convention. Like all international tribunals, it is not concerned with the separation of powers within the member state. When it says that a question is within the margin of appreciation of a member state, it is not saying that the decision must be made by the legislature, the executive or the judiciary. That is a matter for the member state."

Later in his speech, Lord Hoffmann discussed the reasons that courts of this country should normally follow Strasbourg jurisprudence, Then at paras 36–38 he said this:

"… But none of these considerations can apply in a case in which Strasbourg has deliberately declined to lay down an interpretation for all member states, as it does when it says that the question is within the margin of appreciation.

37. In such a case, it is for the court in the United Kingdom to interpret articles 8 and 14 and to apply the division between the decision-making powers of courts and Parliament in the way which appears appropriate for the United Kingdom. The margin of appreciation is there for division between the three branches of government according to our principles of the separation of powers. There is no principle by which it is automatically appropriated by the legislative branch.

38. It follows, my Lords, that the House is free to give, in the interpretation of the 1998 Act, what it considers to be a principled and rational interpretation to the concept of discrimination on grounds of marital status …