BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

United Kingdom Journals


You are here: BAILII >> Databases >> United Kingdom Journals >> Conjoined Twins: the cutting edge
URL: http://www.bailii.org/uk/other/journals/WebJCLI/2002/issue5/clucas5.html
Cite as: Conjoined Twins: the cutting edge

[New search] [Context] [Printable version] [Help]


 [2002] 5 Web JCLI 

Conjoined Twins: the cutting edge


Bev Clucas

Lecturer in Law, University of Hull
e-mail: B.R.Clucas@hull.ac.uk

and

Kath O'Donnell*

Lecturer in Law, University of Hull
e-mail: K.J.ODonnell@hull.ac.uk

* The authors would like to acknowledge the helpful observations of those who have commented on earlier versions of this article, and in particular the comments of the anonymous referee.

 

Copyright © Clucas and O'Donnell 2002.
First published in the Web Journal of Current Legal Issues



Summary


Does medical possibility entail legal and ethical necessity? Continuing advances in medicine prompt this question, which has been brought sharply into focus in the last few years by the issue of the separation of conjoined twins. Is separation morally and legally permissible if one life may be saved at the expense of another? The Court of Appeal has claimed to provide a definitive answer (at least for incompetent children) in Re A (children) (conjoined twins) [2000] 4 All ER 961 but we contend that much of the reasoning in the case is flawed. The case exposes the shortcomings of the welfare principle, and side-steps the crucial importance of explicit and inescapable moral and theoretical legal argument. We argue that it may be possible to justify separation in some circumstances and we provide a coherent framework of moral and legal reasoning in order to do so, starting with conceptions of law. Our application of this framework involves a morally objectivistic legal idealism, based upon Gewirth's Principle of Generic Consistency.


Contents

Introduction
1. A cautionary point
2. Separation questions
2.1 Does being conjoined make a difference?
2.2 The nature of conjoined twins
2.3 Morality and legality
2.4 How far can examination of domestic and international positive law
determine the legal status of the sacrifice of a conjoined twin?
The three levels of inquiry.

2.5 Whose moral theory, and why?
3. Re A (children) (conjoined twins):
3.1 An overview of the case
3.2 The Court of Appeal judgments
3.3 The welfare principle
4.1 A court of law, not of morals?
5. A practical framework
5.1 Level three: the concept of law
5.2 A question of morality
5.3 The criteria for moral argument
5.4 Levels One and Two of our inquiry
6. Conclusion
Bibliography



Introduction


Much of the history of humankind has been taken up with the advancement of the possible, and the present day is no exception. What is novel is that the extensions of what it is possible to do, from more efficient ways of killing people to the cloning of humans, have fast outstripped the ability of society to come to a consensus on what is permissible or right. Some of these novel questions on the cutting edge of ethics are addressed by committees of specialists and experts, such as the Human Fertilisation and Embryology Authority in relation to assisted reproduction, but in many cases the issues fall to be decided by the courts. One current and particularly poignant debate is that of the separation of conjoined twins where both children will die unless one child is sacrificed to save the other.

Conjoined twins exist on the margins of our notions of embodiment and individuality. They challenge the boundaries of medical, ethical and legal possibility (and permissibility) and their existence poses a threat to entrenched social values about the worth of lives that differ from the norm of one individual, one body (Dreger 1998, Myser and Clark 1998). High profile instances of the sacrificial separation of conjoined twins, from twins in Philadelphia in 1977 (see Annas 1987) to Re A (children) (conjoined twins: surgical separation) [2000] 4 All ER 961, have highlighted the fact that separation decisions seem to be reached on a case-by-case basis on their perceived merits, which does nothing for the internal coherence of the reasoning in the case – judges may agree on outcomes but for different reasons, which presents a problem in the application of precedents – or for the case's coherence within the law – a deserving outcome in one case may cause tensions in related law. As is often said, 'hard cases make bad law'.

When separation is considered, for example on the grounds of medical necessity, we accept that a choice has to be made between competing arguments. The leading UK case on separation, Re A, made such a choice. The determining factor of that case was the court's interpretation and application of the welfare principle. In this article, we argue that the welfare principle was inappropriate for the determination of issues of this kind: neither the content of 'welfare' nor the application of welfare arguments in competition situations has been settled or fully explored. We contend that the judgments in Re A leave much to be desired, and do not even begin to approach the decisive status to which they aspire. We maintain that it is possible to attempt an answer to the dilemma of separation of incompetent children, but only by employing a rigorous, comprehensive and coherent argument that explores the relationship between law and morality, and operates from the starting-point of an objective moral theory. The relationship between law and morality must be explored because rival conceptions of law have different relationships with morality, and thus may result in different outcomes on a given question. Whatever moral theory is employed needs to be objectively justified if it is to establish its mandatory applicability to a question. We cannot claim to offer a solution that will streamline the law, but suggest that it is important first of all to focus on the substantively right decision before being concerned with uniformity of the case-law.

This paper is in five parts. After an admonition in the first section, in the second we highlight certain assumptions about separation, and then consider questions that need to be addressed. We identify three levels of inquiry (the application of rules; reasoning where no directly applicable rule exists; and conceptual questions) which a thorough consideration of separation requires. Not all of the issues that arise from separation can be approached in a paper of this size: we restrict ourselves to welfare (other questions are raised by Re A: for the criminal issues, see Huxtable 2002, McEwan 2001, Michalowski 2002, Tausz and Smith 2001 and Uniacke 2001; for the human rights issues Black-Branch 2001 and Clucas and O'Donnell 2001; and for the question of who decides see Clucas and O'Donnell 2001 and Michalowski 2001). We limit ourselves to separations similar to Re A. The third section of this paper is taken up with a discussion of Re A as the leading UK case on conjoined twins, but does not seek to be a definitive analysis of all the legal issues raised. The penultimate section examines Ward LJ's claim in Re A that the court was a court of law, not of morals. The final portion of the paper is the most ambitious, and outlines and attempts to implement a logical framework for the resolution of ethical issues, with a consideration of the three levels of inquiry previously identified. Our framework is one of legal idealism founded on moral objectivism, and for this we owe an obvious debt to Alan Gewirth, Deryck Beyleveld, Roger Brownsword, and Shaun Pattinson. Our conclusions on separation in the instance of Re A match those of the Court of Appeal, but our reasoning is very different: the right decision was reached for the wrong reasons. We contend that the process by which an outcome is reached is at least as important as the outcome itself in any given case, and even more important when the reasoning from one case is applied subsequently. We aim to provide a practical and rigorous framework for the resolution of other cases of this kind.

 

Top | Contents | Bibliography



1. A cautionary point


In the publicised cases to date, there has been a presumption on the part of the doctors and judges involved that where it is possible to separate conjoined twins with a good chance of at least one survivor, this ought to be done (Sheldon and Wilkinson 1987, Thomasma and Muraskas 1996). We concur with Annas (2001, p.1108) that '....once the twins are separated verbally it is only a matter of time before they are separated surgically': it becomes assumed that separation ought to take place. We believe that at least where there is no medical necessity to separate, the impetus for separation results from societal constructs of what is acceptable (Myser and Clark 2001 p 47; Dreger 1998). Robert Walker LJ's discussion in Re A about restoring to Mary her bodily integrity ([2000] 4 All ER 961, at 1066 and 1069) provides a striking example of the puissance of these social constructs: conjoined twins are seen as not 'normal', therefore they ought to be made to resemble the norm. For example, during the writing of this paper, there was much speculation over the fate of conjoined twins Natasha and Courtney Smith who were born to a Hertfordshire couple, in April 2002. They shared a heart and liver, and the parents and doctors were agreed that they should be separated. The heart, which was situated a little further to Natasha's side of the conjoined body, would be allocated to Natasha, who was described as the dominant twin. Natasha was thus selected as the survivor and Courtney as the sacrifice. The doctors and the parents seemed to share the assumption that separation should take place (despite the lack of success in previous cases of shared hearts), an assumption arguably reinforced by the highly publicised decision in Re A . Separation was subsequently decided against after a revised opinion that neither twin would be likely to survive.

The motives driving this argument may be exemplary – twins may be separated in order to enjoy a conventional life and to avoid discrimination against their difference (although in Mary's case, separation merely made her fit a conventional coffin) – but we are unconvinced that fears for twins' welfare justify such a step. Chang and Eng Bunker, the original Siamese twins, appear to have lived full and happy lives despite their difference, as do other conjoined twins (Raffensperger 2001). In centuries past women have suffered serious discrimination on the grounds of their sex, but we think no one would argue sincerely that women ought to have been killed, or to have been surgically altered in order to pass as men, to spare them this trauma. Where there is discrimination, this must be knocked down rather than shored up. Separation is never without risk, whether medical or psychological, and prejudice, albeit well-meaning prejudice, must not play a part in the decision. We query the urge to make conjoined twins conform to societal norms. For societal norms to be adequate criteria of moral permissibility, an argument must be adduced that establishes them as such. We are not convinced that this is possible. In section 5 below we give our account of an argument for moral permissibility that is based on an argument for necessary truth rather than conventional wisdom.


2. Separation questions


The impetus for judicial consideration of separation tends to be perceived medical necessity. The court's dilemma is most acute when the rights or best interests of one twin are perceived to be in conflict with those of the other (of which more below). The paradigm of the separation dilemma is Re A, where both twins were expected to die within a few months (a few years at most) if remaining conjoined. One had a very good chance of independent ('normal') life if the twins were surgically separated, which operation would cause the death of the other twin. We examine Re A in more detail in section three below. We cannot hope to address all the questions implicated in separation, and we restrict ourselves in section 5 to the rights and wrongs of the separation of conjoined twins where the survival of one twin by separation is an alternative to proximate death for both.


2.1 Does being conjoined make a difference?


If a person has rights by virtue of being a person (whatever this may mean), then a person who is conjoined will also have rights. Are conjoined twins one entity with one bundle of rights, or are they two distinct persons? If they are distinct persons, may there be an important difference in the rights that it is intelligible for a conjoined person to exercise? Physical possibilities that exist for singletons – such as privacy – become much more problematic for conjoined twins, as it is not possible to be conjoined and physically alone. For example, if we accept that a right to privacy exists, to speak of a right to privacy in the sense of a right not to be with or observable by another person against one's will is nonsense in the case of individual conjoined twins. Other rights might have to be reconsidered also.

It is important to note that even if the content of a conjoined right changes – by this we mean the particular instantiation of a right in specific circumstances – this does not affect the weight of the right in question. We are used to variety in the instantiations of a right to autonomy – for example, X's right to prefer tea and Y's right to prefer coffee when they both desire a hot drink – but may need to remind ourselves that variety obtains with respect to other rights whose content usually seems fixed.


2.2 The nature of conjoined twins


Conjoined twins are monozygotic twins, where a fertilised egg has begun to divide, but the process has stopped before total division occurred. Discussion of conjoined twins frequently begins by exploring the extent to which conjoined twins are one person or two (e.g. Sheldon and Wilkinson, 1997). Perhaps the more fundamental debate is the extent to which personhood depends on or is distinct from embodiment (Myser and Clark, 1998; Wasserman, 2001; Munro, 2001; Kaveney 2001), although Finnis comments negatively on the prevalence of the mind-body dichotomy (1993, p334). We content ourselves here with pointing out that conjoined twins can be seen as inhabiting an intermediate category of embodiment between one and two, yet exist in a society which seems unable to contemplate with equanimity the degree of blendedness and interrelatedness which conjoined twins exhibit, or the idea that such intermediate categories may exist or need to be constructed. The current rules by which society is regulated, being focussed on individualised rights, have difficulty in accommodating these categories of interrelatedness and interdependence (Sheldon and Wilkinson 1997 p151; Munro 2001).

Conjoined twins exist outside society's perception of the normal, and are frequently described in terms which emphasise their abnormality, or suggest that their development is not natural (the judgments in Re A provide an excellent example, and see also the acute deconstruction of television documentaries by Myser and Clark (1998)). The intuitive response is often that conjoined twins are two individuals whose development did not proceed 'properly': that their conjoined state is a mistake which ought to be corrected: that their normal development should be achieved by other, surgical means. Hence, conjoinment is seen as a medical condition that requires treatment, and the indicated treatment is invasive surgery to complete the process of splitting that was halted in the womb. Conjoined twins do deviate from the norm, but the language of abnormality tends to bring perjorative implications: deviation from the norm does not necessarily require that conjoined twins be reshaped or reconstructed to better fit the norm.

The preference for separation is based on assumptions that being different is not in children's interests. But it is clear that (at least some) adult conjoined twins are happy in their conjoined lives. They feel that conjoinment does not bar them from experiencing aspects of 'goods' of human existence, such as marriage and parenthood. They see no need for separation in adulthood: their conjoined lives are normal to them. Adult conjoined twins who have not been separated are known to express their contentment with their lives and their desire to remain conjoined (Raffensperger 2001). If we regard conjoinment as a developmental abnormality that should be corrected, then how much respect are we showing to the lives of those who are born and live conjoined? We may be unable to comprehend or accommodate their embodiment, but that does not justify our 'carving' them to fit into our mould (Myser and Clark, 1998, p47). An emphasis on the appropriateness of separation simply reflects our inability to accept disability and difference and to concede that there may be equal value in lives that exist outside our conceptions of the normal.
Our perceptions of a child's physical and emotional well-being (how rights to or interests in well-being are instantiated) are constructed around certain normative values. Conjoined twins exist outside many of those norms. It seems plain that we cannot apply the same normative standards of well-being to conjoined twins as to singletons.


2.3 Morality and legality


If answers to separation questions have any pretensions to definitive status (and answers to separation questions ought to have pretensions to definitive status, as persons' well-being and lives may be radically altered by them), it should immediately be evident that they must seek moral and legal justification.

Morality and moral principles in general, are to be understood as being located in a particular place within practical reason (reasons for action). Moral principles have the following characteristics: they are justificatory, that is, they provide reasons for action rather than excuses; they are aimed at guiding the actions of the recipients of the principle; they are addressed to persons other than the originator of the principle; they are concerned with the rights or interests of those other than the originator of the principle, and they are categorical, that is, obligatory irrespective of the desires of the recipient of the precept, and having a force that overrides non-moral principles. (For further discussion of the delineation of morality, see Gewirth 1978, pp 1-3 and Beyleveld and Brownsword 1994, p 126.) A moral justification will provide a reason why a course of action ought or ought not to be followed. We contend that whether or not a proposition is labelled 'moral' is irrelevant: if it aims to be justificatory, action-guiding, other-addressing, other-regarding and categorical, then it is a moral proposition. In 5.3.b below, we provide an additional reason for subsuming separation questions within moral discourse: it will be seen that if a question is within the sphere of practical reason (reason for action), it is subject to a supreme moral principle, and an action may be morally permissible, morally impermissible, or morally optional. In sum, both the characteristics of judgments made about separation, and the situation of separation questions within practical reason, make separation judgements subject to moral scrutiny (whatever this may entail).

With regard to law, for the moment it will be enough to define law as 'the enterprise of subjecting human conduct to the governance of rules' (Fuller 1969, p 96). What precisely this entails will hinge upon the nature of the connection between law and morality. Legal positivists will deny that there is anything other than a contingent connection between law and morality, whilst legal idealists will be committed to the view that there is a conceptually necessary connection between law and morality. For the moment we do not enquire what, if anything, is required by that enterprise (this is taken up in section 5, below). In the limited present sense of 'law', an answer to questions of separation must be justified in law because the law consists of the rules that regulate our lives.


2.4 How far can examination of domestic and international positive law determine the legal status of the sacrifice of a conjoined twin? The three levels of inquiry.


Legal inquiry on the subject of sacrifice comprises different levels of investigation. The first and second levels will be the most familiar to lawyers who confine their scrutiny to the rules that are posited for enforcement in the polity in which we live, but long-term acquaintance is not the only, or the best criterion, to delimit our study.

The first level of inquiry consists in an exploration of the rules. Are there rules that prohibit the intentional killing of a conjoined twin? If a rule exists that has direct application to the case in hand, then an answer to the positive legal status of sacrifice of conjoined twins can be found easily. How much influence is had by international law and convention untertakings becomes merely the technical question of positive legal effect.

If, however, there is no clear rule concerning the killing of conjoined twins, then the second level of investigation must be commenced. If there are rules concerning the killing of (non-conjoined) persons, do these rules apply in an identical fashion to conjoined twins? Lawyers will be familiar with the techniques of argument by analogy, which could be used in support of an equal prohibition on the killing of conjoined persons as non-conjoined persons. However, the capacity to suit one's arguments to one's conclusions, the ability to argue towards a pre-determined goal, is of no assistance when we are in doubt as to the answer.

Do rules against homicide apply in the same way to conjoined twins as to other persons? A legal positivist may consider the law to have 'run out' when there is no direct rule to apply to a particular circumstance. In such a case, a judge must use her discretion to make law. The answer to the question 'what does the law say about the sacrifice of a conjoined twin?' becomes whatever a judge says it is. In contrast, Ronald Dworkin (1998), who can be understood as a type of legal idealist, denies emphatically that the law can ever run out: there are legal principles (albeit not always express principles) that lie behind the rules, which, on careful study and interpretation, will yield the answer. A judge, according to Dworkin, may find the law, but never has the strong discretion to make the law. It can be seen that the question of what is the law in a hard case where there is no clearly relevant rule leads one on inexorably and logically to the third level of investigation: the concept of law.
The debate over the concept of law – what is law – does not only become a live issue in a hard case, but this is not always immediately apparent to the non-lawyer or the practitioner or the non-theoretician. When one is faced with rules enforced by political power with the label 'law', it is all too easy to assume that the label fits the conception. (We do not consider all of the points related to the concept of law debate here: there is, for example, the important question of what obligation there is to comply with an unjust rule. See Beyleveld and Brownsword 1994 for discussion of this, and many other questions.) It should be clear, however, that even if one is used to conventional acceptance of what is law, this concept of law nettle must be grasped once conventional rules give no clear answer. One's conclusion to the concept of law question – whether there is no necessary conceptual connection between law and morality (legal positivism) or whether there is a necessary conceptual connection between law and morality (legal idealism or natural law) – will determine one's approach to the law, and the law in a hard case (See sections 4 and 5 below).


2.5 Whose moral theory, and why?

The question of separation, where one conjoined twin is killed to save another, is a moral as well as a legal question (see 2.2, above). Whether one takes the view that a judge must make a (moral) decision when the law runs out, or that the law cannot run out and is based around a moral principle or set of principles, one must inevitably inquire as to whose (moral) principle should determine the outcome, and why should this particular theory be used. If there is no way of adjudicating between competing (moral) principles, there can be no right answer or even preferable answer when asked should we separate conjoined twins. We resume on this point in section 5, below.


Top | Contents | Bibliography



3. Re A (children) (conjoined twins):


3.1 An overview of the case


With these questions in mind, we turn to the apparently definitive answer provided by the Court of Appeal in Re A (2000). The judgments exemplify the tangle of ethical and legal issues involved in sacrificial separation, and illustrate the potential for inconsistency which flows from what seems to be result-oriented decision-making. We summarise key aspects of the case here but we are not seeking to provide an exhaustive analysis of all the points on which the judgments can be criticized: for detailed commentary on the judgments, see Clucas and O'Donnell (2001), Uniacke (2001), Harris (2001), McEwan (2001), Michalowski (2001, 2002) Hewson (2001), Wasserman (2001), Annas (2001) and Gillon (2001).

The case concerned conjoined twins identified as Jodie and Mary (real names respectively Gracie and Rosie) who were born in August 2000. Their parents, Michelangelo and Rina Attard, lived on the Maltese island of Gozo and had come to the UK, to St Mary's Hospital in Manchester, for medical expertise that was not available in their home country. The twins were joined at the pelvis but had separate vital organs, with the main exception of the liver and bladder. Due to the nature of the join between them it was relatively easy to regard each child as possessing a separate heart, lungs, two arms, two legs etc, but the twins shared a circulatory system. Their circulation was joined at the main artery, and crucially, Mary's heart and lungs were not functional. The shared artery meant that the heart attributed to Jodie was supplying oxygenated blood to both babies. The court (and most writers) had no difficulty in allocating body parts between the twins and throughout the judgments (and commentaries on them) the organs are referred to as belonging to one or other baby. The functioning heart is thus described as Jodie's, rather than as a shared organ. The ways in which the twins are described and the consequences of such description are significant.(1)

The medical evidence was clear. Mary's heart and lungs were not functioning: she had not been able to breathe for herself and had she been born a singleton, she would not have survived birth and could not have been resuscitated. In addition, her brain function was significantly impaired. Jodie's heart was sustaining Mary, but it could not do so infinitely. As the twins grew, the strain of supporting the conjoined body would lead to heart failure, and would cause other complications before death. It was anticipated that when Jodie's heart failed, she would die first, followed shortly by Mary. If the twins remained joined, then both would die, within a period of time estimated as between six months to a few years. If the twins were separated, medical opinion was that there was an excellent chance that Jodie would survive, and, with reconstructive surgery, could lead a relatively 'normal' life. However, separation would mean clamping and severing the shared circulatory system and as a result Mary would die. Although the question of placing Mary on life support was raised at an earlier stage, none of the doctors or judges involved treated this as a realistic possibility, undoubtedly influenced by the degree of brain damage present (although McEwan questions some of the reasoning here: 2001, pp 254 and 258).

Separating the twins from each other would preserve Jodie's continued life, but would prematurely end Mary's. On the other hand, leaving the twins conjoined would cause the proximate deaths of both. The other possible option of separating if an emergency situation developed was discounted as posing too high a risk to Jodie and diminishing the likelihood of a successful outcome for her. The case therefore starkly encapsulates the ethical and legal dilemmas involved in sacrificial separation, and the questions outlined earlier.

At first instance, handling the application as an emergency hearing, Johnson J had held that Mary's life would be short and hurtful to her and to prolong it would be 'very seriously to her disadvantage' (cited by Ward LJ, [2000] 4 All ER 961 at 988). Mary's interests were paramount and separation would be in her best interests. He affirmed that if separation were a positive act, not an omission, it would not be lawful and could not be made lawful, however, he held that separation would be an omission not an act. He decided that separation would be an interruption or withdrawal of the blood supply, analogous to the withdrawal of feeding and hydration in the Bland case, and would therefore be lawful (Airedale NHS Trust v Bland [1993] AC 789).


3.2 The Court of Appeal judgments


The Court of Appeal, consisting of Ward, Brooke and Robert Walker LJJ, delivered more considered judgments, which all concluded that separation would be lawful, albeit for different reasons.

The judges took the view that they must first ask whether they could override the parents' refusal of consent, a question which they sought to answer by the use of existing principles on consent and the welfare principle. They then considered whether Mary's death would be an unlawful killing, i.e. whether it would be intended and whether there would be any applicable defence. The case thus ranged across a number of principles central to issues in medical, family and criminal law, but the judgments are not unimpeachable (although we do not consider all the issues involved in this paper).

All three judges agreed that they were bound by the welfare principle in s 1(1) of the Children Act 1989 and the idea of best interests: that the court's paramount consideration was the welfare of the children. Ward and Brooke LJJ held that the best interests of Jodie and Mary were in direct conflict, but that this could be resolved by balancing the benefit to Jodie against the harm done to Mary. They held that the benefit to Jodie outweighed the harm to Mary, who would die in any event when Jodie's heart failed, and thereby overruled the parents' refusal to consent. By contrast, Robert Walker LJ agreed with Johnson J at first instance, holding that the separation was in the interests of both children.

As to whether Mary's death would amount to an unlawful killing, there was less agreement. Ward LJ held that her death would be intended but would not be murder because of quasi-self-defence. Brooke LJ held that Mary's death would not be murder because there was a defence of necessity. Robert Walker LJ held that Mary's death would not be intended.

We consider all three judgments to be flawed. We make the following criticisms, which we will support in detail in the sections that follow. Not only are the judgments mutually contradictory, but they are also internally inconsistent in many places. They are clearly based upon a range of values (particularly values about conjoined or separated life) which were neither argued nor properly justified. Their treatment and application of established principles of law can be strongly criticised and in their ostensible hope to justify their desired outcome, they overturn established principles of criminal law without sufficient argument or reason. They claim that there is no place for moral reasoning in the resolution of this complex (and undeniably moral) dilemma, but the judgments make clear use of certain moral arguments, in the absence of a properly reasoned framework.

Referring back to three levels of enquiry which we outlined previously in section 2.4 above, it is clear that, at the first level of enquiry, there are no rules which specifically relate to conjoined twins nor to the killing of one to save another. At the second level of enquiry, there are rules that relate to the medical treatment of non-conjoined children, and to the killing of non-conjoined persons. Could these rules and principles be used to resolve the issue before the court, namely the treatment or killing of conjoined children? The court quite clearly claimed that this was the level of inquiry with which it was engaged: 'This court is a court of law, not of morals, and our task has been to find, and our duty is then to apply the relevant principles of law to the situation before us - a situation which is quite unique' (Ward LJ [2000] 4 All ER 961 at 969). However, it is equally clear that they were in fact engaged in consideration of moral issues: see 2.3 above.


3.3 The welfare principle


Our discussion from hereon will focus on welfare and the balancing of the twins' best interests. The Court of Appeal began its search for applicable rules and principles by turning to the principle which underpins the modern law relating to children: the paramountcy or welfare principle. When the court's inherent jurisdiction is invoked, its first and paramount consideration is the well-being, welfare or interest of the child (Re B (A Minor) (Wardship: Sterilisation) [1987] 2 All ER 206). In addition, the Children Act 1989 contains the statutory formulation that 'when a court determines any question in respect to (a) the upbringing of a child .... the child's welfare shall be the court's paramount consideration' (s1(1) Children Act 1989). This essentially means that this is the court's only consideration (Freeman 2001, p 276). The rhetoric of welfare has thus become the touchstone of almost all areas of the law dealing with children. However, we argue that the application of the welfare principle fails to resolve the dilemma of whether sacrificial separation is morally and legally permissible. The welfare test is not an appropriate or sufficient mechanism for resolving competing rights to life; nor should the right to life be subordinated to a subjective principle of welfare (and see Harris 2001, p 229).


3.3.a The meaning and content of welfare


The concepts of 'welfare' and 'best interests' invariably govern cases concerning children. Significantly, these concepts, which are never clarified, are presented as objectively determinable and shared norms.

Although Ward LJ purports to discuss the meaning of the welfare test, his discussion of the meaning of best interests is limited to two short quotations ([2000] 4 All ER 961, both at 994) from recent authorities which do nothing to address the content of welfare or the way in which the rhetoric of welfare is arguably used to validate or rationalize a desired outcome.

'Best interests are not limited to best medical interests.' (Butler-Sloss LJ in Re MB (an adult: medical treatment) [1997] 2 FCR 541 at 555)

and

'In my judgment best interests encompass medical, emotional and all other welfare issues.' (Dame Elizabeth Butler-Sloss P in Re A (medical treatment: male sterilisation) [2000] 1 FCR 193 at 200).

The content and limits of 'best interests' are never clearly defined. As Dewar, among others, points out, the concept of welfare is not susceptible to legal definition with the clarity or objectivity normally associated with (or claimed by) legal principles (1992, p. 381), but crucially it is usually discussed by judges as if it is, and as if that definition is known and shared. We suspect that the language of welfarism is a rhetoric which enables a system to protect and promote those values to which it wishes to adhere most strongly. That rhetoric has long been exposed to criticism on the grounds of its indeterminacy and dependence upon the decision-maker's own value judgments (Mnookin, 1975, p.226, and see Reece, 1996, as an example). These value judgments are frequently revealed through the judicial use of language, as can be seen in Re A itself. Here, the judgments are clearly influenced by perceptions of 'normality', and the value (or lack of value) of a conjoined life.

Conjoined twins provide an excellent example of the problem associated with the use of the welfare principle: our perceptions of children's well-being are constructed around certain normative values. Conjoined twins exist outside many of those norms and it may be that those normative standards of well-being cannot necessarily be applied to conjoined twins. The Court of Appeal, however, judged the twins' welfare and well-being entirely by reference to standards of the normal existence and development of non-conjoined persons.


3.3.b The application of the welfare principle to Jodie and Mary


Jodie's best interests

The application of the welfare principle to Jodie seems relatively unproblematic. Jodie's life expectancy, joined to her twin, was short, between a few months or a few years. An elective separation held a high chance of survival for Jodie (around 94%), and a good prognosis for Jodie's post-operative life, although it was possible that she would need further surgery. A life along the lines of the worst outcome would have inconveniences and unpleasantness, particularly in view of the remote situation of the parents' home. There was no indication that her intellectual development would be other than normal and it was anticipated that there would be no adverse psychological consequences resulting from the separation. By contrast, an emergency separation would carry much greater risks for Jodie. If the twins remained conjoined, the strain of providing a blood supply for both herself and her sister would ultimately cause Jodie's death through heart failure: Jodie would predecease Mary, thereby bringing about Mary's death. Postponing any separation in the hope of Mary's death occurring first and an emergency separation taking place at that point was therefore not felt to be a viable option.

Considering only Jodie's interests, elective separation seemed the best option, and Johnson J found that:

in medical terms Jodie's life would be virtually as long as and would have the quality of that of any ordinary child... For Jodie separation means the expectation of a normal life...(cited in Ward LJ's judgment, [2000] 4 All ER 961, at 988).

The implications of this view are clear: if separated from her twin, Jodie has a good chance of a 'normal' life, and therefore separation is in her best interests.

Ward LJ, considering the appeal against Johnson J's judgment, considered that:

Far from being plainly wrong, Johnson J was in my judgment plainly right to conclude that the operation would be in Jodie's best interest. (ibid., 996).

Ward LJ's judgment placed considerable emphasis upon the quality of Jodie's post-operative life and the relative normality that would be achieved, and both Brooke LJ and Robert Walker LJ agreed.

Mary's best interests

It was accepted by all the judges in the case that the outcome of the operation to separate would be Mary's death, although there was some uncertainty about whether that outcome would be 'intended' in the context of criminal culpability.

Johnson J refers to Mary's state as 'pitiable' (ibid., 988) and his description of Mary's condition is couched in emotive language. The medical witnesses could not be certain of the extent to which Mary registered either pleasure or pain. Despite this, and despite evidence that Jodie would be unable to learn to move around while attached to Mary, Johnson J was influenced by the prospect of Mary being in pain as Jodie began to gain mobility and dragged her around. He concluded that:

'...the few months of Mary's life if not separated from her twin would not simply be worth nothing to her, they would be hurtful... To prolong Mary's life for these few months would in my judgment be very seriously to her disadvantage' (ibid., 988).

He therefore held that separation would be in Mary's best interests.

Although Ward LJ disagreed with Johnson J's findings and conclusion here, his judgment is phrased no more neutrally: '(i) I agree with the judge's assessment: one pities Mary because her position is utterly dire for she exists pathetically on borrowed time' (ibid., 997).

Ward LJ holds (ibid., 1004) (and Brooke LJ agrees) that separation is not in Mary's best interests:

The only gain I can see is that the operation would, if successful, give Mary the bodily integrity and dignity which is the natural order for all of us. But this is a wholly illusory goal because she will be dead before she can enjoy her independence and she will die because, when she is independent, she has no capacity for life. The operation is not capable of ensuring any other improvement to her condition or prevent any deterioration in her present state of health. In terms of her best health interests, there are none. To be fair to the hospital, they do not pretend that there are (ibid., 998).

In a stark contradiction, the obvious inconsistencies in Johnson J's judgment did not prevent Robert Walker LJ from agreeing with his conclusions and finding that the proposed operation, which he agreed would be an invasive act which would end her life, would be in Mary's best interests:

even if it were assumed that Mary is no more capable of feeling pain or discomfort than she is of any pleasant sensation or emotion, it is hard to see any benefit to her from continued life (ibid.,1057).

Robert Walker LJ also clearly laid great stress upon the bodily integrity argument. Despite Ward LJ's earlier comments about this being a 'wholly illusory goal', Robert Walker LJ held that 'the operation would give her, even in death, bodily integrity as a human being' (ibid., 1070). This is a highly paternalistic and value-laden decision which implies that Mary's 'right' to a separated body outweighs her interest in her own continued existence.

What all the judges fail to accept is that Mary and Jodie already have bodily integrity: they have a conjoined body (see Watts, 2001, Harris, 2001, Hewson 2001). The judges identify conjoined bodies as lacking in both integrity and dignity, clearly only able to conceive of integrity and dignity as belonging to separated bodies. They fail to treat the twins with the respect that they insist elsewhere (e.g. [2000] 4 All ER 961, 966 and 1054) is due.

The value of welfare

There can hardly be a clearer illustration of the shortcomings of the welfare principle than the judgments in Re A. Two judges (Ward LJ and Brooke LJ) held that the separation was not in Mary's best interests, since it would cause her death. Yet, using exactly the same principle, the two remaining judges (Johnson J at first instance and Robert Walker LJ in the Court of Appeal) held that separation – i.e. death – was in Mary's best interests, and one leading commentator (Freeman, 2001) agrees. What is considered to be in Mary's (or Jodie's) best interests depends purely upon the values placed upon the factors under consideration. The use of the welfare principle to resolve life or death dilemmas cannot be justified if it is so subject to the vagaries of judicial values and preferences.


3.3.c When best interests conflict: the limits of paramountcy


Re A exposes a further flaw in the welfare principle: the emphasis on paramountcy means that it is incapable of resolving conflicts between the welfare of different children. The court is concerned with two children, whose interests appear to conflict. The welfare of both cannot be paramount.

By holding that separation was in Mary's best interests as well as in Jodie's, Johnson J at first instance and Robert Walker LJ in the Court of Appeal managed to neatly sidestep the difficult issue of balancing the interests of two children, both of whose welfare is paramount under the Children Act s1(1). By refusing to distort their interpretation of Mary's best interests, Ward LJ and Brooke LJ, however, had to deal with the contradiction. In the absence of statutory rules, the Court of Appeal turned to two previous cases that had addressed this problem.

Birmingham City Council v H [1994] 2 AC 212 concerned a contact application by the 16 year old mother of a 1 year old child in care. In that case, the House of Lords avoided the contradiction inherent in s1(1) by holding that only the welfare of the baby was paramount, since only the baby was the subject of the application before the court. This clearly could not be applied to the situation in Re A, so Ward LJ and Brooke LJ turned to the Court of Appeal judgments ([1993] 1 FLR 883). The Court of Appeal had decided that the welfare of both mother and baby were in issue and fell within the paramountcy principle. The Court had argued that where the interests of the children were each to be treated as paramount but were directly opposed to each other, the court must conduct a balancing exercise and reach the decision which represents the least detrimental alternative overall. In the words of Evans LJ in the Court of Appeal in the Birmingham case, the court must 'have regard to potential detriment for one in the light of potential benefit for the other' (ibid., 899).

This line of reasoning was followed in Re T and E (proceedings: conflicting interests) [1995] 1 FLR 581. Here, the Court of Appeal was required to decide upon residence applications for the three children of a family, whose interests conflicted. The court again took the view that it must balance the interests of the children against each other to achieve the least harm overall.

Ward LJ therefore sought to carry out such a balancing exercise between Jodie and Mary. He stressed that Mary was incapable of independent life; that had she been a singleton she would have been stillborn or died shortly after birth; that she was in any event doomed to die within a few months or years at the very most. By contrast Jodie's position was very different: she was capable of independent life; it was anticipated that, if separated, her independent life would be of a good quality and reasonable duration; but if not separated, she would inevitably die as a result of her physical attachment to Mary. In short, Mary could not be saved from a proximate death, but with a high degree of certainty (and with a prognosis of a near normal life), Jodie could. The harm to Mary – i.e. death, which could only be postponed and not averted – was held to be clearly outweighed by the benefit to Jodie of continuing and near-normal life. On this basis, both Ward LJ and Brooke LJ held that the separation was justified, although they distinguished this from the question of whether it could lawfully be carried out without incurring criminal liability (see 3.4 post).

The Court of Appeal concluded that Jodie's (potential) long term continued life was worth more than Mary's inevitably short life. However, it is clear from the way in which the girls were discussed that this valuation was not simply a quantitative one based on their relative life expectancies: the judgments repeatedly refer to the near normal quality of life which Jodie is expected to enjoy, as contrasted with the poor quality of life which Mary enjoys and would continue to enjoy. The balancing exercise which has been carried out depends upon value judgments about the quality of life, which detract from the absolute nature of the right which is at stake: the right to life.


3.3.d The value of life


At the heart of Re A lies the proposition that the life of one person morally should, or legally may be sacrificed to save the live of another. The court made general statements about the fact that Mary and Jodie were both living persons possessed of equal rights to life ([2000] 4 All ER 961, e.g. at 1001 and 1026), but Ward LJ's analysis of the way in which life should be valued, which is the most comprehensive, draws heavily upon Airedale NHS Trust v Bland [1993] AC 789 and specifically upon John Keown's subsequent (1997) exposition of the possible approaches to valuing human life. We argue that the court's interpretation of the sanctity of life approach resulted in an appraisal of the relative worthwhileness of the twins' lives (in which Mary lost), rather than an assessment of whether the treatment proposed was worthwhile (which is what the court purported to judge).

Before considering the coherence of the sanctity of life approach and Keown's distinctions generally, some mention must be made of the applicability of Bland to Re A. As Ward LJ recognised, Mary was not receiving treatment in the sense that Tony Bland was:

Mary is not receiving treatment (or any substantial treatment) at the present time. Such care as she receives in hospital will of course prolong her life but there is no question of withdrawing that care or that treatment. What is under consideration is the active invasion of her body. That will not prolong her life. It will terminate it. ([2000] 4 All ER 961, at 1003).

If Bland as a factually analogous precedent is unsuitable, is its reasoning on the sanctity of life applicable to the twins? It was necessary in Bland to distinguish between evaluating the worthwhileness of treatment and the worthwhileness of the patient's life: their Lordships were quite clear in stating that to say that Anthony Bland's life was not worthwhile would be to 'cross the Rubicon', on the other side of which lies euthanasia. If life were not sacred, then it would be permissible to kill those whose lives were deemed not worth living.

The sanctity of life is one of three approaches on a continuum identified by Keown (1997): vitalism, 'sanctity of life' and 'Quality of life' (his capitals). Keown defines vitalism as the position that human life has an absolute value that must be preserved or lengthened at all costs. 'Sanctity of life' is characterised as the view that human life has intrinsic dignity that protects it from unjust attack: the right to life is essentially a right not to be intentionally [and unjustly] killed. Sanctity does not mean vitalism: rather than the preservation of life at all costs, Keown argues that the sanctity doctrine accepts that 'in many cases it is perfectly proper to withhold or withdraw life-preserving treatment.... The sanctity principle holds that there can be no moral obligation to administer or undergo a treatment which is not worthwhile' (ibid., 485). He distinguishes the question of the worthwhileness of treatment from the worthwhileness or value of the patient's life: the latter assessment is argued to be the third approach of 'Quality of life', which focusses on the worthwhileness of a patient's life rather than the merits of the treatment. 'Were one to engage in judgments of the latter sort, and to conclude that certain lives were not worth living, one would forfeit any principled basis for objecting to intentional killing' (ibid., 485). Keown claims that it is the principle of the sanctity of life on which English law (with the anomalous exception of Bland) is based (support for this was most recently and explicitly provided in the Pretty cases in the House of Lords and the European Court of Human Rights, [2001] UKHL 61 and 35 EHRR 1 respectively).

It is critical to Keown's reasoning on the sanctity of life that decisions regarding medical treatment must focus on whether the treatment is worthwhile, not whether the life is worthwhile. Worthwhileness of treatment is explained in the following way:

A treatment may not be worthwhile either because it offers no reasonable hope of benefit or because, even though it does, the expected benefit would be outweighed by burdens which the treatment would impose, such as excessive pain. (Keown 1997, 485).

Keown illustrates the distinction between worthwhileness of treatment and worthwhileness of life with reference to two hypothetical babies:

...Angela, a baby born with Down's syndrome and an intestinal blockage. Her doctor informs her parents that the blockage can be removed by a straightforward surgical operation and that, if it is not so removed, Angela will die. The doctor and parents, judging that the treatment will clearly benefit Angela by saving her life while involving only minimal burdens, such as the usual discomfort associated with the operation, decide that the operation would be worthwhile or proportionate and should be performed.

'Contrast Angela with Bertha, a baby born with a terminal illness which will inevitably lead to death in a matter of hours. Her doctor informs her parents that, due to respiratory difficulties, she may stop breathing at any time and asks whether they would like attempts at artificial ventilation to be made should her breathing falter. Given that such efforts could not hope to reverse Bertha's inevitable decline, and might impose significant burdens on her, they decide against ventilation as it would be disproportionate. (ibid., 485-6).

Had the judgements in these two scenarios been made on the basis of a 'Quality of life' approach, then neither Angela nor Bertha might have received treatment, on the basis that a life with Down's syndrome and life with terminal illness were not worth living.

Keown criticises the reasoning in Bland on the grounds that a purportedly sanctity of life approach slips into a 'Quality of life' approach. For example, tube-feeding was identified as 'treatment' in the case. Quoting Dr Keith Andrews (ibid., 492), Keown points out that tube-feeding is extremely effective because it fulfils its purpose: to provide sustenance for the body. Any judgment that condoned the withdrawal of tube feeding would have to be made on some other ground: for example, that Tony Bland's life was not worthwhile. (Keown's disapproval of the reasoning in Bland was so profound that he concluded 'Bland rendered the law morally and intellectually misshapen' (ibid., 503).) This elision between 'sanctity' (legitimate) and 'Quality' (illegitimate) of life approaches is maintained in Ward LJ's summing-up of their Lordship's reasoning in Bland: 'vi) Since there was no prospect of the treatment [tube feeding] improving his condition [pvs] the treatment was futile...' ([2000] 4 All ER 961, at 998, our insertions in square brackets).

Ward LJ discusses Bland at some length (ibid., 998 et seq.), and is aware of academic criticism of the case, including Keown's. It appears throughout the majority of his discussion of Mary in relation to 'sanctity of life' and 'Quality of life' arguments that he has taken Keown's comments on board. He uses Keown's words: the question is that of the worthwhileness of the treatment, not the life. However, in his conclusion to how the balance between Jodie and Mary's interests should be struck, we find the following passage:

(ii) The question which the court has to answer is whether or not the proposed treatment, the operation to separate, is in the best interests of the twins. That enables me to consider and place in the scales of each twin the worthwhileness of the treatment. That is quite a different exercise from the proscribed (because it offends the sanctity of life principle) consideration of the worth of one life compared with the other. When considering the worthwhileness of the treatment, it is legitimate to have regard to the actual condition of each twin and hence the actual balance sheet of advantage and disadvantage which flows from the performance or non-performance of the proposed treatment. Here it is legitimate, as John Keown demonstrates, and as the cases show, to bear in mind the actual quality of life each child enjoys and may be able to enjoy. In summary, the operation will give Jodie the prospects of a normal expectation of relatively normal life. The operation will shorten Mary's life but she remains doomed for death. Mary has a full claim to the dignity of independence which is her human entitlement. In the words of the Rabbinical scholars involved in the 1977 case in Philadelphia, Mary is 'designated for death' because her capacity to live her life is fatally compromised. The prospect of a full life for Jodie is counterbalanced by an acceleration of certain death for Mary. The balance is heavily in Jodie's favour. ([2000] 4 All ER 961, 1010).

In this decisive passage, Ward LJ forgets that he has already dismissed the separation operation as not treatment for Mary – a fact that makes the balancing of treatment for both twins impossible. He is correct that one must have regard to the actual condition of each person, but only insofar as it is necessary to tell whether the treatment, being directed at the condition (this is not the case for tube-feeding regarding the improvement of pvs), would improve the condition (be worthwhile). When Ward LJ talks of 'the dignity of independence which is her human entitlement', a claim to which is incidentally nonsense on the grounds of being impossible for Mary to realise whilst being alive, he makes the shift to the valuation of the individual twins' lives, the step he regards as illegitimate.

Ward LJ's analysis of the right to life and sanctity of life is heavily influenced by his perceptions and values about the conjoined state. The story he tells is one where an egg fails properly to divide, resulting in conjoined rather than separated twins. In this story, conjoined twins are a mistake of nature, and more specifically, we can see that it is Mary who is to be regarded as the mistake. Mary should have been born as a physically separate child, or so the reasoning runs: had she been born separated, she could not have survived; she only survives because of her shared connection with Jodie. Her life is 'unnaturally supported' (ibid., 1011). By extension, it seems that Mary should not be alive. For Ward LJ, Mary has the right to life, but she has no right to be alive, especially as it is that very connection which will also cause Jodie's earlier death.

Mary may have a right to life, but she has little right to be alive. She is alive because, and only because , to put it bluntly, she sucks the lifeblood out of Jodie. Mary's parasitic living will be the cause of Jodie's ceasing to live. If Jodie could speak, she would surely protest, 'Stop it, Mary, you're killing me. Mary would have no answer to that (ibid., 1010) (although Uniacke replies on her behalf: 2001, p 212).

Annas also offers hypothetical conversations between the twins, which might be more altruistic.

Each twin might also, of course, consider the other twin to be an integral part of herself, a view that would preclude separation. Any of these hypotheses is plausible, but made up monologues cannot take the place of legal analysis' (2001, p 1104).

Both Brooke and Robert Walker LLJ have the same ability to mould the right to life into seemingly contradictory shapes. Both judges take the 'Mary as mistake' idea to its ultimate conclusion: that Mary's right to life is to a life in a form which they recognise and value: separated and not conjoined ([2000] 4 All ER 961, at 1061, 1066). Thus, separation will give Mary the bodily integrity and dignity which they clearly believe to be lacking in the conjoined state (ibid., 1052, 1069). As formulated by the judges, this appears to be valued over and above Mary's right not to be deprived of her life (and see Freeman 2001). Mary's life, and her right to it, it is clear, is of virtually no value: she is 'self-designated for a very early death' ([2000] 4 All ER 961, Brooke LJ at 1051).

Throughout the case it was argued that the correct approach to the valuation of life is the sanctity approach: a right not to be intentionally deprived of life, rather than the vitalist approach, which would include the preservation of life. If this approach were rigorously applied it would in fact lead the court to the opposite outcome. What is proposed is the perservation of Jodie's life at the cost of taking another life. The strict application of principles of sanctity would appear to hold that sacrificing Mary's life in order to ensure the continued preservation of Jodie's is morally impermissible. The emphasis on worthwhileness of treatment, which is used to support the claim that sanctity of life does not require the preservation of life at all costs, is turned on its head in order to support the claim that Mary's continued existence is not worthwhile, while Jodie's is. Jodie's life has (through the consideration of welfare) been valued more highly than Mary's, largely because it can more readily be reshaped to conform to the norm.

Our conclusion on Keown's reasoning differs from others: see e.g. Harris (2001, p. 225), where he writes:

Both Ward LJ and Brooke LJ approve of the following authoritative version of the distinction [between the value of the treatment and the value of the patient's life] by Lord Goff in the Bland case: '[T]he question is not whether it is in the best interests of the patient that he should die. The question is whether it is in the best interests of the patient that his life should be prolonged by the continuance of this form of medical treatment or care.' The distinction is equally spurious in both the versions attributed to Keown and to Lord Goff because it articulates a distinction without a difference.

It is indeed 'astonishing that anyone should entertain this supposed distinction for a moment' (ibid). We agree with Harris that Lord Goff's formulation is a Quality of life judgement by the back door. But it is equally astonishing that a reading of Keown (1997) could be thought to be advocating Lord Goff's approach.

It is not surprising that Ward LJ makes this shift from sanctity of life to Quality of life once he has committed himself to 'balancing the worthwhileness of the treatment'. The only way in which his dilemma can be resolved in Jodie's favour is by preferring one child's life to the other. Keown's sanctity of life approach cannot have any other result than being opposed to separation in the case of Jodie and Mary: entirely the opposite conclusion to that of Ward LJ and Bland (and see Beyleveld and Brownsword (2001, 262-3) for a different discussion of Ward LJ's reliance on Keown). Our conclusion may seem similar to Ward LJ's, but in fact is significantly different: our application of the Principle of Generic Consistency, below, is a precautionary analysis of the relative moral status of Jodie and Mary, and not an evaluation of the quality of their lives. The evidence by which we judge non-arbitrary moral status leads us to prefer Jodie's rights to Mary's rights in a situation of unavoidable conflict: see section 5 below.


Top | Contents | Bibliography



4.1 A court of law, not of morals?

We identified earlier the strange proposition of the court, that in deciding on the twins' separation, it was not deciding on a moral (as opposed to a non-moral) issue. Ward LJ, in his leading judgment, was determined to stress the point that 'this court is a court of law, not a court of morals' ([2000] 4 All ER 961, 969).

4.1.a Does 'morality' equate to religion?

This claim by Ward LJ that the court is a court of law and not of morals, seems in many ways an odd statement to make. Discussion of causing the death of a person seems to be one of the clearest instances of moral discussion. Perhaps though, the word 'moral' has restricted connotations for Ward LJ. Perhaps by 'morals' he really meant 'religion'. If this is the case, then we find no fault with his declaration that the court is not a religious court and that the case should not be decided by reference to specifically religious principles.

4.1.b An attempt to exclude evaluation of the decision?

Perhaps Ward LJ meant to exclude any possibility of evaluating the moral rightness or wrongness of the court's decision. This is a much more contentious enterprise. Either the debate on the separation of the twins was a moral debate (concerned with providing a decision for or against separation that was justified and categorical; aimed at guiding the actions of the persons involved, and concerned with the interests or rights of the twins), subject to a supreme moral principle (see section 5 below), or it was not. If the debate was, as it is evident, the type of debate that is conceived of as 'moral', then it matters not what label Ward LJ wished to give it. He cannot define his way out of criticism of the moral substance of the court's decision.

4.1.c Ultra vires

One of the strangest consequences of Ward LJ's declaration that the court is a court of law and not morals is to exclude the court's competence to decide a moral debate. This debate is a moral as opposed to a non-moral debate: the court claims not to have the competence to decide a moral debate. The court, by its own admission, exceeded its jurisdiction in pronouncing on the fate of Jodie and Mary.

4.1.d A conceptual claim: there is no necessary connection between law and morality

There seem to be two more likely interpretations of Ward LJ's words: the first relates to a particular theoretical position in the concept of law debate considered here; the second could be merely an indication that the court was concerned with a decision that was controversial as to the substantive moral content, considered in 4.1.e below.

To make the strong claim that a court is a court of law and not of morals is to deny that there is any necessary conceptual connection between law and morality: it is to espouse the separation thesis of H. L. A. Hart (1994). We propose to give a fuller account of our position in this debate below. For the moment, it is sufficient to outline what must be proved if this interpretation of Ward LJ's statement is to be anything other than arbitrary.

To establish that there is no necessary conceptual connection between law and morality, one must prove that a problem of social order (law being the enterprise of subjecting human conduct to the governance of rules) need not necessarily presuppose the governance of an overarching moral principle (see Beyleveld and Brownsword 1994, especially chapters Three and Four, for discussion of valid methodologies). Merely to assume the superiority of legal positivism, as Ward LJ's statement seems to do, is to make an arbitrary claim. We are not obliged to agree with this claim until it is proven. If legal positivism is mistaken, and legal idealism is the superior theoretical position, then the legal validity, as well as the moral validity of the judgment is called into question.

If it was indeed Ward LJ's intention to make the claim that there is no necessary conceptual connection between law and morality, perhaps part of his motivation was to avoid the duty on the court to (at least) attempt to come to the correct moral decision – a duty required by morally objectivistic legal idealism, and one popularly linked to the idea of legal idealism or natural law theory in general. A duty to (attempt to) arrive at a correct moral decision is bound to be problematic if a judge's conception of a correct answer to a moral question equates to a mere conviction, as seems to be the case with Ward LJ.

4.1.e A controversial moral decision

It may be that Ward LJ did not intend to make a claim about the concept of law, but instead wished to present the court's deliberations as musings on a (non-moral) legal question in order to avoid controversy about whether the court reached the substantively correct moral position. This reading seems to be in conflict with his express conviction of having made the right decision ([2000] 4 All ER 961, at 998) that we characterise as a moral rather than non-moral conviction. However, it may be that Ward LJ meant simply that this was the right legal decision on a moral issue. This position is a) unconvincing (see 4.1.b, above), and b) reduces to a stand for legal positivism when carried to its logical conclusion (see 4.1.d, above), but we need not suppose that this is what Ward LJ intended.

Moral convictions and moral principles habitually are regarded by those who possess them as self-evident unproblematic truths, until there is conflict between rival positions. For the majority of people, it is simply obvious that murder is morally wrong, and since no mainstream moral principle opposes this view, the grounds for its claim to veracity remain unchallenged. But when one moves from almost universal convictions to more intractable moral problems, the issue of why a particular principle ought to be accepted comes to the fore. The case of the conjoined twins is one such instance of moral controversy, because of the inevitable death of Mary resulting from separation, and the inevitable (yet preventable) proximate death of Jodie without separation.

If the court were to claim a) that as a court of law, it had a duty to come to a substantively correct moral decision, and b) that it had done so, it would undoubtedly find itself under attack, because others would dispute that the decision to allow separation was the correct moral decision. Unless the court has a satisfactory means of demonstrating that it has good reasons for thinking its decision is correct or at least preferable to alternatives, such a bold claim would meet with derision. If Ward LJ simply intended to distance himself and his colleagues from this undesirable situation, then this is understandable, even if not defensible (for he had not established that the court was not under a duty to come to the right moral decision).


Top | Contents | Bibliography


5. A practical framework


Re A is not the first case concerning conjoined twins, and Courtney and Natasha Smith remind us that it will not be the last. Although it is to be hoped that that their separation would also have been brought before the court, the confused reasoning in Re A does not make us sanguine that future cases will necessarily be decided by reference to a sound framework of legal or moral principle. McEwan's (2001) concern about the implications of Re A is well-founded. Despite the court's opinion that this case was unique, and that their judgment was confined to tightly defined circumstances, it has already been followed, in the Australian case of Queensland v Nolan ([2001] Q.S.C. 174). The health of the conjoined twins in that case was deteriorating and the judge seems to have followed Robert Walker LJ's reasoning in holding that the operation would be in the weaker twin's best interests (Grubb, 2002).

We propose the following framework as the appropriate method for resolving the questions we identified in section 3 above, and as a guide for decision-making in the case of conjoined twins. In contrast to the usual approach identified in section 2.4 above, we begin with level three, and work in reverse order. This is because a proper answer to the question 'what is the law on x' can only be given after conclusions have been reached on the concept of law, and any relevant moral principles. This is the case whether legal idealism or legal positivism is espoused. The technical lawyer's focus on the rules first and other points second, if at all, can be characterised as either an attempt to answer this question on the assumption that legal positivism prevails, or as an attempt to answer the question that fails as a proper attempt (because the concept of law is not adverted to at all).

We contend that a proper attempt to answer the question 'what is the law on the separation of conjoined twins' must proceed in the following way: by examination of the concept of law debate, via moral philosophy, and then to consideration of what rules can properly be called 'law' (what rules are rightly posited and enforced). Our answers to these questions employ morally objective legal idealism or a version of natural law that resonates with the principle 'lex injustia non est lex', but we maintain that those with rival views on the concept of law must still employ our framework, that is, begin with the conceptual foundations of level three of our inquiry, in order properly to attempt an answer.

The following sections address the concept of law; whose moral theory ought to be employed and why; problems of application of our preferred moral theory to children; solutions to this problem; application of the moral theory to conjoined twins; and consideration of other competing rights to be taken into account regarding separation.


5.1 Level three: the concept of law


One's answer to the concept of law question – whether law is or is not necessarily conceptually connected to morality – will determine one's answer to questions of law. The question of the nature of the intersection between law and morality is perhaps most obvious to non legal theorists when determining the law in a hard case, but in fact the concept of law question is logically prior to any determination of 'the law' in a case: what counts as law is dependent on one's conception of law.

In this paper, we adopt a legal idealist conception: that law is necessarily conceptually connected to morality. More specifically it is a morally objectivistic legal idealism: the correct conception of law is law as morally legitimate power, and moral legitimacy is measured by an objectively justifiable moral principle. For this purpose, we adopt Gewirth's Principle of Generic Consistency (hereafter 'the PGC', discussed in detail below), which we believe is an objectively justified moral principle.

Law is the enterprise of subjecting human conduct to the governance of rules (Fuller 1969, p 96). To suppose that there is the necessity of subjecting human conduct to the governance of rules is to presuppose a problem of social order. To presuppose a problem of social order is to conceive of human nature and the nature of human interactions in a particular way. More specifically,

...we contend that there are a number of human interests and factors which, as a constellation, define the human social condition insofar as this condition is capable of presenting a problem of social order. In arguing for a Natural-Law Theory [morally objective legal idealism] we contend that the defining factors are a number of facts of human finitude: such as, human vulnerability and mortality; limited understanding, reason and knowledge; limited time to pursue goals; and limited resources; and that the defining interests are:

Personal Interests. These divide into two categories:

Biological or Life Interests. This category includes interests in life, health, comfort, food, sex, fun, knowledge, and control of nature. Such interests are common to most human beings, and are functions of the fact that human beings are purely biological creatures.
Purely Individual Goals. These are interests which derive from differences between individuals, rather than from factors which human beings share in common.

Moral or Reciprocity Interests. This category includes interests in mutual accountability or responsibility for action, mutual care or concern for the welfare of persons, respect for persons, and empathic knowledge.

Autonomy Interests. These are manifested by a desire to pursue other interests actively (Beyleveld and Brownsword 1994, pp.121-122).

Only if it is assumed that human beings act in pursuit of these constellated interests can we conceive of human relations resenting a problem of social order that needs to be regulated. To have to presuppose that moral interests form a necessary part of the constellation of interests that have to be presupposed in order to conceive of a problem of social order, is to have to presuppose a necessary conceptual conncetion between law and morality. That it is necessary to presuppose moral interests as part of the constellation of interests is, in our view, demonstrated by Alan Gewirth's argument to the PGC (see 5.3.b below). Gewirth's PGC is not only the supreme moral principle: it is the supreme principle of practical reason (reasons for action), and overrides other non-moral and lesser moral principles. Legal reason (law being the enterprise of subjecting human conduct to the governance of rules) is a sub-category of practical reason, and is therefore governed by the PGC. (For a much fuller explanation of the relationship between law and morality see Beyleveld and Brownsword 1994 Chapters 1-4 for discussion and methodology, particularly pp. 120-145 from which the above account is derived).

In determining whether the separation of conjoined twins is legally permissible, it must first be ascertained whether the particular proposed separation is morally permissible (with reference to the PGC). If separation is morally impermissible, one would expect separation to be legally impermissible. On the other hand, if separation is morally permissible, it may be either legally optional or legally required (depending on whether to require separation would be an infringement of other rights derived from the PGC). Our paper confines itself to legal permissibility or impermissibility, and does not attempt to answer the question of the scope of legal regulation (that is, whether all actions subject to moral regulation ought to be subject to legal regulation, or only some. The former position would make all immoral actions illegal; the latter need not).

It should be emphasised that Beyleveld and Brownsword's definition of a legal system, which we employ (a system that subjects human conduct to the governance of morally legitimate rules), does not restrict legal systems to formal systems comprising legal professionals as HLA Hart's (1994) definition does (see also Adams and Brownsword 1999, Chapter One). A legal system is any system in which human conduct is subject to governance by rules. We see no difficulty in designating as legal systems a community (such as a religious community) or an institution (such as a university) or a family, provided that these are social situations where human behaviour is subject to rules. Such a legal system will be a valid legal system when its rules do not violate the PGC. A person may therefore be simultaneously the subject of many legal systems arranged in a hierarchical and/or distinct manner, all of which systems must comply with the Principle of Generic Consistency.


5.2 A question of morality


The first point that must be acknowledged in a serious attempt to resolve the issue of separation is that this issue is a moral issue. As we have indicated above, it is irrelevant whether the terminology of 'moral problem' is used: it is impossible to conceive of separation without conceiving of a problem as defined in 2.3, above. It is also impossible to exclude a practical problem from the sphere of the PGC: see below.

Separation involves, prima facie, the rights of two children, and the rights of others, such as the parents. What grounds (if any) have we for believing that any of the parties involved has rights? If they do, how would separation impact upon all these rights?


5.3 The criteria for moral argument


We are convinced by Gewirth's argument to the PGC (see below), and so obviously employ his moral principle in determining moral questions, but it is possible to employ a different moral principle at level three of our enquiry (for example, a utilitarian principle, or a rights principle based on interest theory, rather than our preferred will-theory rights principle). It is our contention that it is only possible for a will-based rights theory to give a persuasive answer to this question 'why ought I to be moral?', in addition to a consideration of what conditions need to be fulfilled for a persuasive answer.


5.3.a Reasons for a moral theory


One of the characteristics of a moral principle is that it provides justification for action. The search for an answer to the dilemma of conjoined twins requires just that: a justificatory principle that guides us to right action. Yet a moral principle requires something in addition to this: the principle itself must be justified in order to sustain its authority in the face of disagreement with the principle. To have authoritative value, the justification of the principle must be sound not only from the point of view of those who agree that one ought to be moral but disagree that this principle is the right one, but also from the point of view of those who assert that there is no reason at all to be moral. It would be naïve to hope that one principle be accepted by all persons. What must be aimed at – and what is more satisfying than mere conventional acceptance – is an argument that demonstrates that a) one must be moral, and b) what 'being moral' requires.

The search for a proof of morality is controversial. Some people will deny that it is possible to prove such a thing. Others may claim that philosophers who have attempted this have failed. Many more people may simply have a prejudice against proofs or truths outside science. It is the received view in the current intellectual climate that absolutism is naïve.

If a diagnosis be necessary, it may be that this conception of knowledge results from the intellectual swing away from the view of the world as created by God, to one in which there is no purpose or design, but only organisms that have survived owing to their match to the environment. But moral absolutism, and in particular the theory that we recount below, should not be dismissed as deism clothed in post-Enlightenment argument, or as a by-product of intellectual shortcomings. If one is to attempt a solution to the dilemma of conjoined twins, one must also make a good faith reading of a theory that claims to provide, or be the means of providing, right answers to moral questions and reasons for these.


5.3.b Alan Gewirth


Among the philosophers who claim to have provided a proof of morality, we believe there to be one particularly strong candidate: Alan Gewirth (1978). His supreme moral principle is the Principle of Generic Consistency: 'act in accord with the generic rights of your recipients as well as of yourself', loosely, always act in such a way as to respect the things that make it possible for yourself and others to act at all and act successfully. Gewirth's argument is controversial and has received criticism from many quarters, as well as responses to those criticisms (see e.g. Regis 1984, Beyleveld 1991, Kramer and Simmonds 1996, Singer 2000, and a response to the latter two by Beyleveld, forthcoming in Ratio Juris).

Gewirth (1978) claims to have provided a transcendental proof of why we must act morally. By this it is meant that the Principle of Generic Consistency (PGC) is a truth which is conceptually necessary as a condition of our experience. Because of our existence, and the manner in which we exist, we cannot deny the veracity of the principle. We are, like you, the recipients, or envisaged recipients, of precepts concerned with action. Such practical precepts may be, but do not need to be, moral in content. There are essential, pervasive, generic features of action which apply whether the precept is moral – 'thou shalt not kill' – or instrumental – 'use a hammer rather than a screwdriver to drive nails into wood' – to hypothetical – 'if you wish to avoid points on your driving licence, you ought to avoid breaking the speed limit'. Voluntariness and purposiveness are the generic features of action. Voluntariness and purposiveness must be the generic features of action (all action, not just moral action: if the latter were the case, the argument would beg the question) if practical precepts are to be relevant at all: it is pointless and illogical to direct precepts at persons who are incapable of guiding their action by these precepts.

Gewirth's argument is dialectical, rather than assertoric. It must be undertaken from the internal viewpoint, as an internal conversation, and only holds good from within this internal viewpoint. It is not an argument for establishing concrete, external truths, or as Brownsword phrases it,

the argument does not make a direct 'assertoric' claim that the PGC corresponds to the moral furniture of the universe (nor, of course, that it is, in fact, generally accepted by agents) (Brownsword 1993, p 257).

If the PGC categorises murder as wrong, then it does so 'only' (!) in the sense that it proves, necessarily and logically, that an agent must regard murder as wrong.

The summary of the argument below is phrased in a manner that aims to be accessible. We distinguish the dialectical statements with speech marks. Our summary is not the most precise formulation of the argument to the Principle of Generic Consistency, and should not be taken as such. For this, and a comprehensive defence of the argument, see Beyleveld (1991).

Imagine that you are the 'I' of the argument.









The steps in the argument to this point have established that 'I must regard myself and others as having the generic rights.' The move to considering that all other agents must regard themselves as having the generic rights and also must assent to the PGC is only clear when the Argument for the Suffciency of Agency has been understood. The argument takes the form of a reductio ad absurdum, in which the proposition 'I must consider that I have the generic rights on the sufficient ground that I am an agent' is proved by taking as a premise a statement which negates this, and then showing that a contradiction occurs when this negation is taken in conjunction with the previously established steps of the argument:

'I must regard that I have the generic rights by virtue of being an agent and for no other reason.' If I wish to negate this, I do so by supposing that 'I have the generic rights on other grounds, perhaps because I teach law.' I could, then, claim that 'I have the generic rights by virtue of being a law teacher.' But to hold this is equivalent to saying that 'if I were not a law teacher, I would not have the generic rights.' There is an obvious contradiction here: the argument to the PGC has already established that it follows logically from necessary premises that I must claim that 'I am an agent, therefore I must regard myself as having the generic rights.' It is clear that I must consider that 'I have the generic rights simply because I am an agent,' and for no other reason.

Since I regard that 'I and all other agents have the generic rights,' I must accept the Principle of Generic Consistency (PGC): 'respect the generic rights of your recipients as well as of yourself.' (The generic rights are rights to freedom and well-being.)

Universalising logically, I must consider that 'as I am an agent, and I must accept the PGC by virtue of being an agent, therefore all other agents must also accept the PGC, because they are also agents.'

Those who are unconvinced by the dialectical argument to the PGC might find more persuasive Beyleveld's (1996) argument that a (contingent) comittment to human rights is a comittment to the PGC.


5.3.c Applying the Principle of Generic Consistency to non-ostensible agents


The Principle of Generic Consistency is claimed to be the supreme moral principle that binds all agents to respect the generic rights of all other agents. This requires that an agent respect the things that make it possible to have purposes and generally fulfil one's purposes. There is a hierarchy of goods to which an agent has rights: basic goods, which relate to the prerequisites for action, that is the essential conditions of agency, such as '...life, physical integrity, health and its various contributing factors, general freedom, mental equilibrium and the like...' (Gewirth 1978, 211); non-subtractive goods, which consist of not losing what one regards as good (ibid., 233), that is, relating to an agent's ability to maintain present levels of purpose-fulfilment; and additive goods, which 'consist in the means or conditions that enable any person to increase his capabilities of purpose-fulfilling action and hence to achieve more of his goals...concerned not with maintaining the necessary preconditions and capabilities of one's action but with augmenting and developing these capabilities' (ibid., 240), or in increasing current levels of purpose-fulfilment.

A straightforward application of the PGC would seem to yield a clear result: an operation that has as an inevitable outcome the death of one twin would seem to be the violation of that twin's basic right to life. However, the application of the PGC to children is not so simple, as young children (at least) do not seem to fit the definition of 'agent' (and the definition of an agent is not an arbitrary definition, but is the necessary conception of a being to whom practical precepts may be addressed).

An agent is a being with the capacity for voluntary and purposive action (ibid., 27). Babies are 'debatable cases' (ibid., 75) of agency. It is controversial, to say the least, whether babies and perhaps very young children actually possess the capacity for voluntary and purposive action, in the sense that they are able to reflect on their purposes and decide whether to pursue them. Babies are more likely than not non-agents: it would seem that agents are not bound by the PGC to respect their rights, as babies have no PGC rights (though this does not entail that the PGC requires us to not respect the rights they would have, e.g. to life, had they been agents). At first glance, the absolute moral principle seems to fail to account for our intuitive protection of children (although even if does fail on direct application, it may be that children's rights can be dealt with indirectly, for example as a type of property of their parents, or 'respecting their rights' may be morally optional). If a direct application is the end of the matter, then the PGC can give no guidance in the case of conjoined non-agent twins.

Gewirth circumvents this undesirable state of affairs with his Principle of Proportionality:

When some quality Q justifies having certain rights R, and the possession of Q varies in degree in the respect that is relevant to Q's justifying the having of R, the degree to which R is had is proportional to or varies with the degree to which Q is had. I shall call this the Principle of Proportionality (PP). Thus, if x units of Q justify that one have x units of R, then y units of Q justify that one have y units of R (ibid., 121).

Gewirth's valiant attempt is unsuccessful, however (see Hill 1984 and Gewirth's (1984) response; and see also Beyleveld and Brownsword 2001, pp 117-119 and pp 129-132). Potential for voluntary and purposive action is not enough. Being an agent (having the capacity for voluntary and purposive action, or the capacity to stand in a proactive valuational relationship to one's purposes) is not a capacity that can be had by degrees: either one is capable of valuing one's purposes, or one is not. Nor are the generic rights rights that can be had by degrees: how could half a right to autonomy be coherent? One cannot be half an agent and have half the rights of agency any more than one can half a right to vote at the age of nine. The Principle of Proportionality fails to provide an answer to this dilemma.

As we noted above, the omission of babies and young children from the class of rights-holders derived from the direct application of the PGC does not mean that agents ought morally to kill, maim or neglect non-agent humans. As a will theory with autonomy as its central principle, Gewirth's theory permits agents to bind themselves and their future behaviour, so long as the benefit of the right to autonomy is not waived permanently. This means that agents – and polities made up of agents – may agree to protect, for example by law, the interests of children. Any such legal rights of children would, however, be contingent on agreement between agents, and presumably, those who wish to harm children would not so agree. Child protection would thus be limited to the extent that the freedom of those who wish to harm children can be legitimately restrained by the rights of others to impose their preference (the protection of children) on a morally optional subject matter.

There are other alternative strategies for the protection of children, for example an (empirical) slippery slope argument: to fail to protect young children would lead inevitably to an undesirable state of affairs which would contravene the PGC, for example brutal behaviour to agents; therefore children ought to be protected in order that violations of the PGC be avoided with regard to agents. But arguments of this kind rest on empirical assumptions that are hard to establish or refute. A much better strategy can be found in precautionary reasoning, a particular application of the PGC (but not a modification of the argument to the PGC) developed by Beyleveld, Brownsword and Pattinson (see Beyleveld and Pattinson 2000 and Beyleveld and Brownsword 2001 pp 119-134).

The practical use of the PGC is attended by the age-old philosophical problem of other minds: I know that I am a moral agent because I experience my agency, but how can I know (by external observation) that others are also agents (agency being an internal, subjective quality)? The capacities of voluntariness and purposiveness identified by Gewirth as the necessary and sufficient characteristics of agency are not descriptions of behaviour. When I attempt to judge whether the person with whom I am dealing is really an agent, I make a best guess based on my observations of whether their behaviour accords with what I would expect from an agent, but I can never know that the person is an agent. Ostensibly being an agent (being an ostensible agent) is not proof of agency.

Even though it must be acknowledged that there is no conclusive way of proving that a person is or is not an agent, this is not the end of the story. Even though the truth of the answer to the agent dilemma will remain irresolvable, the moral consequences of the alternative answers to the question are not on a par. If I mistakenly treat Ms Bloggs, the butcher, as an agent, when she is not, I have done no wrong. I have inconvenienced myself by restricting my actions, but I have not violated the PGC, which requires me to respect the rights of other agents. However, if I mistakenly treat Ms Bloggs as a non-agent, and she is an agent, then I have violated her rights and the supreme principle of morality. The PGC has categorical force and requires me to respect other agents' rights. In the second scenario, I have committed a moral wrong, but in the first, I have not.

Beyleveld and Brownsword (2001, pp 121-2) argue not only that actually to violate the PGC in the second type of scenario is wrong, but also that to risk violating the PGC when this can be avoided is wrong:

...under the assumption that the PGC is categorically binding, there can be no justification under any circumstances whatsoever for violating it. Thus, to risk the possibility of violating the PGC, when this can be avoided, is itself to violate the PGC. Therefore, it is categorically necessary to do whatever one can to avoid this consequence (provided, of course, that the actions taken do not conflict with more important requirements to be derived from the PGC) (ibid., 121).

The conclusion is the following Precautionary Principle, which is categorically binding:

If there is no way of knowing whether or not X has property P, then, in so far as it is possible to do so, X must be assumed to have property P if the consequences of erring in presuming that X does not have P are worse than those of erring in presuming that X has P (and X must be assumed not to have P if the consequences of erring in presuming that X has P are worse than those of assuming that X does not have P) (ibid., 122).

Whether or not the consequences are worse is to be judged according to the PGC, in relation to the generic rights violated.

If the being with whom I am dealing is not Ms Bloggs, the ostensible agent butcher, but instead Baby Bloggs, the non-ostensible agent, how ought I to proceed? I cannot be certain that Baby Bloggs is not an agent because my reasoning is fallible. It follows from the Precautionary Principle that I ought to assume that Baby Bloggs is an agent unless the consequences of so assuming are worse than assuming that she is not an agent. Examples of the latter could be where Baby Bloggs and Ms Bloggs both require life-saving medical attention at the scene of an accident, but where the sole doctor to arrive cannot attend to both of them. In this situation the doctor would be justified in treating Ms Bloggs, because, being an ostensible agent, it is more likely that the doctor will violate the PGC (fail to respect an agent's rights) if she neglects the butcher rather than the baby. This can be formulated as a criterion of avoidance of more probable harm:

If my doing y to Z is more likely to cause harm h to Z than my doing y to X (and I cannot avoid doing y to one of Z or X) then I ought to do y to X rather than to Z (ibid., 123, discussed with reference to two partial agents, but equally applicable to our example because the relevant consideration is the likelihood of harm measured by the PGC).

My duties of protection to those who are more probably agents prevail over my duties of protection to those who are less likely to be agents.

The outcome of this discussion of precautionary reasoning is that, although babies and young children (including the conjoined twins of Re A) are not ostensible agents (that is, they exhibit less than comprehensive evidence of voluntariness and purposiveness), because there is some evidence of agent-like characteristics or potential agency – sentience, being human, etc – we must not risk the possibility of violating the PGC in relation to them when this can be avoided, provided that in avoiding this risk we do not violate more important requirements to be derived from the PGC.

In sum, the PGC, which we consider to have been established as the supreme principle of morality, requires all agents to respect the generic rights to freedom and well-being of all other beings that are ostensibly agents. Young children are not ostensible agents. Protection of children is best effected by the avoidance of risk of violation of the PGC in conjunction with precautionary reasoning, which requires that if there is no way of knowing whether or not a young child really is an agent (because of the internal nature of agency), a young child must be assumed to be an agent unless a greater harm (measured by the PGC) would be caused by doing so than by not doing so. We term the protection given to children's well-being by this Precautionary Principle 'children's precautionary rights'. Precautionary rights are not proper PGC rights, but are protections of the (generic) rights children would have if they were agents. In the following section we consider how to deal with the precautionary rights of conjoined twins.


5.3.d Applying the Principle of Generic Consistency to conjoined twins


It is clear from the discussion of precautionary reasoning above that what may be termed children's 'precautionary rights' must be respected unless by doing so we risk violating more important requirements of the PGC – such as failing to respect the rights of ostensible agents.

The case of conjoined twins is more complex than usual conflicts between ostensible agent and partial agent, or even partial agent and partial agent, because the nature of the conjunction between twins may render choices made in respect of one twin a violation of the precautionary rights of the other.

Jodie and Mary, though both partial agents rather than ostensible agents, did not seem do display equally good evidence of partial agency. Whatever else we can say about the empirical identification of ostensible agents, an adult human brain is good evidence of agency. Jodie's mental development seemed to be on a par with other 'normal' babies, and as such would be evidence of partial agency. Mary appeared to have a severely abnormal brain, with a very poorly developed primitive brain and other malformations not compatible with normal development. It is a moot point whether one would class Mary as a potential agent positioned on a further point of the scale of partial to ostensible agency than Jodie, or whether one would take the view that Mary's brain impairment was of such magnitude that she would never approach ostensible agency. In either case, in a scenario of direct conflict of (precautionary) rights, Jodie's rights must prevail because of the greater likelihood that she is an agent than Mary (and thus the greater chance of violating the PGC should we assume that Jodie is not an agent).

If presented with an example of conjoined twins where both were apparently equal partial agents, and both were expected to die shortly because of e.g. a heart condition, but it was possible to choose to save A by separation which would cause B's death, the problem becomes more perplexing. Assuming that the twins are not competent to make the decision themselves (i.e. they are not ostensible agents), the decision-maker has the following options:

Do nothing in the knowledge that A and B will die soon
Choose to separate in the knowledge that this will cause the death of B
Wait until B's condition precipitates a medical emergency and hope that A can be saved

In Re A, option c) was discounted because the outcome of separation was unlikely to be successful for A in the event of an emergency. Unless and until medicine makes c) a realistic option, it can be discounted for practical purposes.

Option b) was the one taken in Re A. Precautionary reasoning would support the outcome of the case simply because of the disparity in moral worth of the two children, but we have excluded this possibility in our present example. In a case where the two twins are, to the best of our knowledge, of equal moral worth, to cause the death of one is equally as bad as to cause the death of the other.

Option a) would seem to many to be the least offensive course: to let both children die would be counter-intuitive when one could be saved, but at least their death would not be caused by any positive action of the decision-maker. However, the moral distinction between act and omission is tenuous when in both cases a choice is made that causes an outcome. We suggest that a doctor who failed to treat a person suffering from pneumonia with antibiotics, in the knowledge that without antibiotics the patient would probably die, has taken something worse than a morally neutral choice. Far from being morally neutral, option a) is to fail to respect the precautionary rights of child A – to save her life when it could be saved – and precautionary reasoning requires us to respect the 'rights' of even an apparently partial moral agent unless the consequences of failing to do so are worse.

The choice of a decision-maker in this situation must be to make the judgment either to fail to save child A where it is possible to save her (a morally culpable decision), or to save A by killing child B (also a morally culpable decision). We cannot say (with any certainty) that one choice is morally preferable when measured by the PGC. A right answer to this question may hinge on the emotional harm done to the agents affected: if the parents would feel more guilty in a particular instance to separate than not to separate, then this could be seen as determining the balance of the considerations involved.


5.3.e Other competing rights


The implementation of children's rights cannot properly be considered in isolation from the rights of their caregivers because of the potential for conflict. Cases on separation ought not to be decided without a reasoned approach to the balancing of any competing rights. What rights do the children's parents or caregivers (hereafter 'parents') have? This question is composed of two separate issues: what impact do the rights of the parents have on the decision (the substantive question), and what role should the parents have in making the decision (the procedural question)? The procedural question will be dealt with in 5.3.f below.

In 5.3.c and 5.3.d, we assumed that the only rights to be weighed in the balance were the rights of the children. The fact that their 'rights' are precautionary rights was irrelevant: their relative moral status was more important (which of the two children whose 'rights' we will inevitably violate by our choices is of lesser moral importance?). When parents are brought into the equation, their superior moral importance as ostensible agents would seem to trump the 'rights' of partial agents. We confine our discussion below to cases of conflict between ostensible agents and those non-ostensible agents who nonetheless have a significant degree of moral status.

It must be recognised that our discussion in this section of the balancing of more important 'rights' of non-ostensible agents with lesser rights of ostensible agents cannot provide a definitive answer. The conflict here involves variables that seem to be incommensurable. However, although it does not seem possible to give a right answer on this application of the PGC, we contend that we still provide good reasons for preferring separation in the case of these conjoined twins.

What rights of the parents are threatened when making the substantive decision? Emotional harm may be done to them if the outcome of the decision is not what they would wish, and freedom from emotional harm is a PGC right. But how is this risk of lesser harm to a more important moral being to be weighed against a certain greater harm to a probably lesser moral being? Let us assume that, in a Jodie and Mary scenario where one twin whose relative moral worth is greater than that of the other, twin J may be separated with a good chance of a 'normal' life, but her parents prefer not to separate (for some reason which does not outweigh the PGC justification for separation). Both options – to separate or not to separate – risk the violation of some rights.

It is our contention that in this circumstance, separation ought to be performed. The PGC right to autonomy does not give agents the right to do whatever they wish – the rights of other agents must be respected. Even though the rights of J are merely precautionary rights, the consequences of being wrong about her partial agency are horrific: to terminate the life of an agent (against her will) is a much greater wrong than to do something that causes an agent distress.

Our qualification in the paragraph above was 'for some reason which does not outweigh the PGC justification for separation'. Is it possible that a graver violation of the parents' rights would outweigh separation? If the rights at stake – to life – were the same, the rights of the superior moral person (the parent or parents) would prevail.

It may be that the parent's right to autonomy will be violated by separation. If the parent of J did not want to spend her life caring for a potentially disabled child, the parent might prefer J to die instead. The basic rights of an ostensible agent would be in conflict with the precautionary basic rights of the child. The ramifications of such a wish are complex and cannot be considered exhaustively here – this would include consideration of the nature of the parent-child relationship and whether a child can be said to be the property of the parent, and the question of whether a parent is estopped from rejecting a child freely conceived – but we must remind ourself that the right to autonomy is subject to the qualification of respecting other agents' rights. A desire not to look after a child does not translate unavoidably into the death of the child.


5.3.f The procedural question: who decides?


What role should the parents have in making a decision on separation? Should the decision rest with the parents or the state, or both?

Agents have duties to themselves and to all other agents. They also have duties to non-ostensible agents which in practice will result in the protection of these precautionary 'rights' (life and physical and mental well-being and development) unless there arises a direct conflict with the rights of ostensible agents. Agents have these duties themselves, and communities of agents have these duties also. In a sophisticated society such as our own, the state can be regarded as the ultimate community in which these duties reside (see below for discussion of communities and legal systems). The state has a duty to protect the precautionary rights of less-than-ostensible agents (direct conflicts with ostensible agents' rights excepting).

The duties owed by the state include the safeguarding of children's positive rights as well as of their negative rights: to respect the generic rights of agents includes positive help as well as non-interference (see Gewirth 1996, Chapter Two). To safeguard children's negative rights would be merely to avoid violating a child's precautionary rights (for example, the state must not kill children). To safeguard children's positive rights would be to prevent the violation of a child's precautionary rights and to help the child secure her precautionary rights when she cannot do so herself (for example, the state must prohibit the killing of children, and ensure that children have adequate food and shelter). Numbered among the different instantiations of the state's positive duty to respect the precautionary rights of children, including the precautionary right to life, must be the duty to prevent the unjustified (by weightier PGC considerations) taking of a child's life.

If the state has a duty to prevent the unjustified taking of a child's life, then the state cannot allow separation decisions – or other decisions that threaten life – to take place without some scrutiny or guidance. This is not to say that all such decisions must be made by the court, though in practice decisions concerning incompetent children would need some kind of reasoned and consistent scrutiny by a state official. The position that Ward LJ highlighted in English law – that if the hospital and the parents in Re A had agreed, separation could have gone ahead without the court's intervention – cannot be countenanced if this duty is taken seriously, nor can the Attards' view that they should be the only competent decision-makers regarding their children. (Whether Ward LJ's statement is correct seems dubious. In the absence of an objection, separation would be likely to have gone ahead where doctors and parents agreed. However, a challenge to this decision could be effected by invoking the wardship jurisdiction of the High Court. In Re B (A Minor) (Wardship: Sterilisation) [1987] 2 All ER 206, Lord Templeman declared that sterilisation of a girl under the age of 18 can only be lawfully carried out with leave of a High Court judge. It is at the very least arguable that separation should be subject to the same safeguards as sterilisation.) Precisely what form of scrutiny or guidance would be necessary is open to debate, although in a situation where the correct outcome is unclear – for example in the hypothetical scenario discussed above in which one of two equally precautionary agents may be saved or both let die – there seems no good reason to remake a decision de novo when all that is necessary is a review of the reasonableness of the decision as measured by the PGC.


5.4 Levels One and Two of our inquiry


In 2.4 above we identified three levels of inquiry. So far in section 5 we have concentrated on the third level, as a resolution of the concept of law debate and which moral principle is to be employed as the basis of our preferred morally objective legal idealism, was prior to a determination of the non-theoretically understood legal issues. We have argued that a decision must be made on the correct conception of law before levels One and Two can be attempted.

Having come to a conclusion on the concept of law debate, levels One and Two of our inquiry can be dealt with together: they comprise jointly the question of adjudication. A legal positivist, having concluded that there is no necessary connection between law and morality, will be content either to apply existing rules on the separation of conjoined twins (level One), or, where there is no clear rule to apply (a Dworkinian hard case), she will follow a theory of adjudication (level Two) which could in principle be any theory that denies a necessary conceptual connection between law and morality. On the other hand, a legal idealist will apply a theory of adjudication that follows from her conception of law.

At this point in our paper we do not wish to become embroiled in the debate over whether Re A can properly be called 'law', or attempt to answer all the questions raised by the adjudication debate. We will content ourselves with the observation that, in our judgment, polities (morally) ought to operate a system of rules that attempts to comply with the PGC, and that judges (morally) ought not to make judgments that violate the PGC. We recognise that some ethical questions (including many instances of the separation of conjoined twins) are so complex that, even if a right answer according to the PGC exists, it may be difficult or even impossible to be sure of attaining that answer. We acknowledge that our attempt to resolve the separation debate may be thought to fall into this category.


Top | Contents | Bibliography



6. Conclusion


As medicine drives us past the limits of our settled moral deliberations, it becomes increasingly important to take stock of the rules and principles by which our lives, and those of others, are ordered. We should be painstaking in our attempts to resolve moral and legal dilemmas that threaten the most basic rights of individuals, such as the separation of conjoined twins. We do not deny that the leading UK case on the separation of conjoined twins contained sincere and thoughtful judgments that at least attempted to resolve the legal dilemma. We maintain, however, that Re A cannot be the last word on separation. A proper attempt to resolve this issue must involve a consideration of embodiment and individuality; a (reasoned) conclusion to the concept of law debate, including a resolution of which moral theory is the basis of a legal idealist conception, and an application of a theory of adjudication that follows from this conception. In our case, we conclude that the correct conception of law is a legal idealism with Alan Gewirth's Principle of Generic Consistency at its heart. We concur with the Court of Appeal with regard to the outcome of Re A, but suggest that the means used to arrive to the Court's decision can be improved upon. Debates on the separation of conjoined twins exist at the cutting edge of law, morality, and individuality. At least let our reasoning be as sharp as the scalpel.


Bibliography


Adams, J. N., and Brownsword., R., (1999). Understanding Law (2nd edition) (London: Sweet and Maxwell).

Annas, G.J. (2001). 'Conjoined Twins: The Limits of Law at the Limits of Life', 344, New England Journal of Medicine, 1104-1108.

Beyleveld, D., (1991). The Dialectical Necessity of Morality (Chicago and London: The University of Chicago Press).

Beyleveld, D., (1996). 'Legal Theory and Dialectically Contingent Justifications for the PGC.', 9, Ratio Juris, 15.

Beyleveld, D., and Brownsword, R., (2001). Human Dignity in Bioethics and Biolaw (Oxford: Oxford University Press).

Beyleveld, D., and Pattinson, S., (2000). 'Precautionary Reasoning as a Link to Moral Action: Proportionality, Precaution, and Duties to 'Agents-in-Part' ', in: Michael Boylan, ed., Medical Ethics (Upper Saddle River, NJ: Prentice Hall).

Beyleveld, D. (forthcoming). 'A Reply to Marcus G. Singer on 'Gewirth, Beyleveld and Dialectical Necessity' ', forthcoming in Ratio Juris.

Black-Branch, J. L. (2001). Being over nothingness: the right to life under the Human Rights Act. E.L.Rev. 2001, 26 Supp (Human Rights)

Brownsword, R (1993). 'Towards A Rational Law Of Contract?' in Wilhelmsson, T (ed.) Perspectives of Critical Contract Law (Aldershot: Dartmouth).

Clucas, B., and O'Donnell, K., (2001). 'Jodie and Mary: a hard case is no excuse.' Conference paper delivered at the Nottingham Law School Health Care and Our Children conference, May 2001. Available from the authors on request.

Devlin, (1965). 'Morals and the Criminal Law' reprinted in Dworkin, R., (ed.) (1977) The Philosophy of Law (Oxford: Oxford University Press), pp.66-82.

Dewar, J (1992). Law and the Family (London: Butterworths).

Dreger, A. (1998). 'The Limits of Individuality: Ritual and Sacrifice in the Lives and Medical Treatment of Conjoined Twins', 29, Studies in History and Philosophy of Biological and Biomedical Sciences, p.1-9.

Dworkin, R., (1998). Law's Empire (Oxford: Hart).

Finnis, J.M. (1993). 'Bland: Crossing the Rubicon?', 109, L.Q.R., 329.

Freeman, M. (2001). 'Whose Life Is It Anyway?', 9, Med. L. Rev., 259-280.

Fuller, Lon L (1969). The Morality of Law (revised edition) (New Haven and London: Yale University Press).

Gewirth, A (1978). Reason and Morality (Chicago and London: The University of Chicago Press).

Gewirth, A (1984). 'Do 'Marginal Agents' Have any Moral Rights?' in Regis Jr., E. (ed.) Gewirth's Ethical Rationalism (Chigaco and London: Chicago University Press) 225-227.

Gewirth, A (1996). The Community of Rights (Chicago and London: The University of Chicago Press).

Gillon, R. (2001). 'Imposed Separation of Conjoined Twins: Moral Hubris by the English Courts?' 27, JME , 3-5.

Grubb, A (2002). 'Conjoined Twins: Re A Down Under', 10, Medical Law Review, 100-102.

Harris, J. (2001). 'Human Beings, Persons and Conjoined Twins: An Ethical Analysis of the Judgment in Re A', 9, Med. L. Rev., 221-236.

Hart, H.L.A. (1959). 'Immorality and Treason', reprinted in Dworkin, R., (ed.) (1977) The Philosophy of Law (Oxford: Oxford University Press) pp.83-88.

Hart, H.L.A., (1994). The Concept of Law (second edition) (Oxford: Oxford University Press).

Hewson, B. (2001). 'Killing Off Mary: Was the Court of Appeal Right?', 9, Med. L. Rev., 281-298.

Hill, J.F. (1984). 'Are Marginal Agents our Recipients?' in Regis Jr., E. (ed.) Gewirth's Ethical Rationalism (Chigaco and London: Chicago University Press) 180-191.

Huxtable, R. (2002). Separation of conjoined twins: where next for English law. Crim. L.R. 2002, Jun, 459-470.

Kramer, M., and Simmonds, N. E. (1996). 'Reason without Reason: A Critiques of Alan Gewirth's Moral Philosophy', 34, Southern Journal of Philosophy, 301-315.

McEwan, J. (2001). 'Murder by Design: The 'Feel-Good Factor' and the Criminal Law', 9, Med. L. Rev., 246-258.

Michalowski, S (2001). 'Reversal of Fortune – Re A (Conjoined Twins) and beyond: Who should make treatment decisions on behalf of young children?', 9, Health Law Journal, 149-169.

Michalowski, S (2002). 'Sanctity of Life - are some lives more sacred than others?', 22, Legal Studies, 377-397.

Mnookin, R., (1975). 'Child Custody Adjudication: Judicial Functions in the Face of Indeterminacy', 39, Law and Contemporary Problems, 226.

Munro, V.E. (2001). 'Square Pegs in Round Holes: The Dilemma of Conjoined Twins and Individual Rights', 10, S. & L.S., 459-482.

Myser, C. and Clark, D.L. (1988). ' 'Fixing' Katie and Eilish: Medical Documentaries and the Subjection of Conjoined Twins', 17, Literature and Medicine, 45-67.

Raffensperger, J. A (1997). 'Philisophical Approach to Conjoined Twins', 12, Pediatric Surgery International, 249-255.

Reece, H. (1996). 'The Paramountcy Principle: Consensus or Construct?', 49, Current Legal Problems, 267-304.

Regis Jr., E., (ed.) (1984). Gewirth's Ethical Rationalism (Chigaco and London: Chicago University Press).

Rogers, J. (2001). Necessity, private defence and the killing of Mary. Crim. L.R. 2001, Jul, 515-526

Sheldon and Wilkinson (1997). 'Conjoined twins, The Legality and Ethics of Sacrifice' 5, Med. L. Rev., 149-171.

Singer, M., (2000). 'Gewirth, Beyleveld, and Dialectical Necessity', 13, Ratio Juris, 177-195.

Tausz, D., and Smith, J.C. (2001). Surgical separation - - whether surgical separation of conjoined twins that would lead to death of non-viable twin lawful. Crim. L.R. 2001, May, 400-405

Thomasma, D. and Muraskas, J. (1996). 'The Ethics of Caring for Conjoined Twins: The Lakeberg Twins', 26, Hastings Center Report, 4-13.

Uniacke, S. (2001). 'Was Mary's Death Murder?', 9 Med. L. Rev., 208-220.

Wasserman, D. (2001). Killing Mary to Save Jodie: Conjoined Twins and Individual Rights', 21, Philosophy and Public Policy Quarterly, 9-14.

Watt, H. (2001). 'Conjoined Twins: Separation as Mutilation', 9, Med. L Rev., 237-245.







[1] The allocation of body parts between the twins implies ownership of those parts, and reinforces claims to exclusivity, rather than equal access to shared resources. For example, in the case of Natasha and Courtney Smith, when separation was discussed, the shared heart which was slightly closer to Natasha's side of the body was described by doctors as 'Natasha's', in what Gillon describes as 'a morally irrelevant locational explanation' (2001:4) . However, when it was decided not to separate, the heart once again was referred to as 'shared'. Whether a functioning heart is to be seen as a shared organ, or described as Natasha' – or Jodie's - heart is more than a question of semantics. The issue of attribution reflects the challenge which conjoined twins pose to our normal understandings of the body, and the consequences of those norms for conjoined twins. Although we have followed the court (and the majority of other writers) in allocating bodily parts to each twin, this should not be taken as an endorsement of the negative attitudes to the conjoined state which we feel are evident throughout the judgments.
(2) If S is a system of reasoning in which 'A has property ð entails that A as property å' then, in S, 'B has property ð entails that B has property å' is also valid.


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/other/journals/WebJCLI/2002/issue5/clucas5.html