English law seems to reflect this dilemma, in sending different signals about the (legal) value of life. This looked like a problem, at least for rational law, but perhaps that is not the whole story. Maybe, at the substantive level, the law is right to ebb and flow as it does, seemingly sceptical of the merits of allowing any one account to dominate. Various commentators do seem to glimpse a compromise of values in the law, if only fleetingly on their journey towards a destination where one value is proclaimed to be the only ‘true’ guide. Thus, for Maclean, the formalities surrounding advance decisions to refuse treatment are ‘perhaps a reasonable compromise between facilitating the uptake of advance directives and protecting vulnerable patients’.1 Lyons, meanwhile, thinks the ruling in MB’s case did a good job of steering a way through the competing opinions of the baby’s parents and doctors, since Holman J seemed to ‘identify a middle ground position. Because of this compromise settlement, perhaps An NHS Trust v MB is a paradigm case’.² To this list, we might add the ruling in the case of minimally conscious M. M’s clinically assisted nutrition and hydration was to remain in place, thereby reflecting the intrinsic value of life: as Baker J put it, ‘the importance of preserving life is the decisive factor in this case’.³ Yet, cardiopulmonary resuscitation (CPR) was not to be attempted if M suffered a cardiac arrest, in view of the ‘significant harm’ that would be caused to M by such an attempt, and the judge left open the possibility of administering antibiotics in the event of an infection.⁴ These limits on the obligation to preserve life might comply with the sanctity of life doctrine, in acknowledging the option to forego ‘burdensome’ interventions, but the judge might have had in mind a different ethic, premised on the instrumental value of life and its concern with avoiding suffering. Here too, then, we might view the courts as seeking to occupy the middle ground.

Not everyone welcomes efforts to find the space between apparently opposing moral positions. The Mental Capacity Act 2005 attempts to preserve the sanctity of life doctrine by insisting that the removal of treatment ‘must not, in considering whether the treatment is in the best interests of the person concerned, be motivated by a desire to bring about his death’.⁵ Coggon has great difficulty with this section, and understandably so, since it (deliberately?) employs terms with no particular legal relevance, seemingly so that it can preserve the veneer of sanctity, without retreating from the instrumental value of life adopted in Airedale NHS Trust v Bland [1993] 2 WLR 316.⁶ This, says Coggon, is ‘a sorry compromise’.⁷

But how ‘sorry’ is this? And should the effort to compromise necessarily prompt an apology? I think not – at least, not always. Rotten compromises might rightly be a source of sorrow,⁸ but principled compromises, which seek to afford space to competing moral perspectives, could be another matter. In this chapter I will mount (and defend) the case for seeking to occupy the middle ground, when certain specific circumstances obtain, and subject to particular constraints.9

6.2 Contemplating compromise

The cases with which we are concerned involve a particular type of conflict, which we may describe as a principled conflict, since it involves a clash of values. The source and nature of the clash might vary in and between particular interests, values, rights and principles;¹⁰ indeed, it might exist in the space between theory and application, as when different advocates of a single principle find themselves in dispute over what it requires of them in practice.¹¹ One and the same person might sense and experience any of these sorts of conflicts, although, mindful of the cases in which we are interested, it is conflict between one person and another (or others) on which we should concentrate.

What we seek is termination of the conflict.¹² There are many routes by which we might arrive at this destination and no fewer methods by which we might travel. A party might accede, concede or supersede; agreement, appeasement or capitulation might ensue; and such outcomes might derive from direct discussion and dispute or, whether by desire or dictate, from the intervention of some third party.¹³ However, not just any resolution will suffice: we need a principled resolution.¹⁴ Moral principles will thus constrain the resolutions reached: for example, a brute exercise of power by one party, by which they secure their victory, will probably not measure up. Law too will provide some of the pertinent principles and will accordingly stake out some of the principled boundaries within which the resolution must be sought.¹⁵ Yet, as we have seen throughout the analysis so far, there may be numerous moral principles within law’s boundaries; as one judge has put it, there are ‘few, if any, absolutes’.¹⁶

We therefore need some way of deciding which moral principle or principles should guide us out of an ethical conflict. Unfortunately, in the conflicts we have surveyed, success cannot be secured by proclaiming the primacy of one set of moral judgments and principles. It is in such circumstances, I suggest, that compromise might be not merely permissible but also morally laudable. This involves seeking a particular type of resolution: a principled compromise, by which compromise is achieved in relation to conflicts of principle, in a manner which is itself also principled.

Understanding of compromise has been advanced by studies in economics, international relations and mathematical game theory but, as Golding appreciated in 1979, the question has been surprisingly neglected in ethical and legal discourse.17 Pennock and Chapman’s collection, in which Golding’s essay appeared, helped shed much-needed light on the issues, but progress has otherwise been slow, save for substantial work by Benjamin,¹⁸ and my own more modest contribution, in a previous book focusing upon the bioethical and legal dilemmas attending euthanasia.¹⁹ In the remainder of this chapter I aim to expand considerably on this work and in so doing suggest why there is merit in contemplating compromise in relation to some of the other dilemmas arising at the limits of life.

Etymologically, compromise refers to an agreement which has been reached through mutual concessions;²⁰ it is, by definition, always suboptimal.²¹ The term can refer to the procedure (process) by which the mutual concessions are secured, or alternatively to the resulting product (solution, end-state, outcome or substance). Kuflik thinks that both features will often be present (to some extent) for each party to a compromise, that is, ‘give-and-take by each (process), resulting in some gain and some loss for all (end state)’.²² The negotiation on which compromise, whether in form and substance, rests can be conducted directly (face-to-face) or through third-party intervention,²³ and it can be consensual or compulsory.²⁴

Many have commented on the ambivalent attitude compromise commands: Kuflik, for example, associates compromise with both ‘moral turpitude’ and ‘moral goodwill’,²⁵ and he suggests that there can be both bad and good compromises.²⁶ Even more positively, Ricœur viewed compromise as ‘a barrier between agreement and violence’, which ‘keeps society from falling apart’.²⁷ Indeed, many commentators describe compromise as a classic feature of democratic societies, in which there is a plurality of people, interests and values.28 As we shall see, there is much to commend this view. But still compromise attracts criticism, hence the common pejorative sense of the word, according to which one’s scruples ‘are compromised’ or one is found ‘in a compromising position’. As Pétrovici noted in 1937, compromise is ‘perpetually condemned in theory and always used in practice’.²⁹ So, what are these theoretical challenges and do they withstand scrutiny?

6.3 Criticising compromise

The disdain with which compromise is often regarded is well captured by Santayana: ‘Compromise is odious to passionate natures because it seems a surrender; and to intellectual natures because it seems a confusion’.³⁰ Here, then, we encounter two standard objections to the idea that compromise can be compatible with morality: that compromise signals a surrender of (moral) integrity; and that, on moral matters at least, to contemplate compromise is to be confused about the nature of morality.

Integrity can be a ‘puzzling’ term,³¹ but the essence of the surrender objection is that the compromiser in some sense undoes himself. As Benditt puts it:

Where principles are concerned, ‘there is an investment of personality’.³³ Benjamin steps in here to argue that we each invest in our own personal world view:

A world view, in turn, structures and grounds a way of life. Integrity, says van Willigenburg, conveys a state of relative wholeness, which requires sufficient consistency in one’s principled commitments and sufficient coherence between one’s principles and one’s actions; or, following Benjamin, one’s world view and one’s way of life. There must also be wholeheartedness, that is, sincere self-direction towards particular commitments, rather than slavishly following one’s peers or passions.35

Holm has criticised the argument, which he attributes to me, that we can compromise (or, in Holm’s words, ‘agree to disagree’) on euthanasia. His attribution may be accurate but, as we shall see, his interpretation of that argument is flawed and there are good reasons to resist his charges. One of his claims is indeed the integrity complaint, since he fears the ‘significant cognitive dissonance and unease’ that compromise will bring.³⁶ Certainly, one can see why (on the preceding accounts) compromise seems such an affront to integrity. But we need not admit the offence.

Kuflik counters the surrender objection by reminding us that there is usually more at stake than the disputed issue itself. The promotion or preservation of peace, the avoidance of force, the importance of open dialogue, and the acknowledgement of reasonable moral diversity will all be important features on the landscape of compromise.³⁷ One’s world view (and corresponding ways of being) will admit of such variety; indeed, as Benjamin says, ‘our identity is constituted in part by a complex constellation of occasionally conflicting values and principles’.³⁸ Sometimes these values will change, as we saw in the case of AE, who at different times aligned herself with Muslim and with Jehovah’s Witness doctrines.³⁹ The self might therefore be seen as a web of commitments, rather than an impermeable core set of values.⁴⁰ Sometimes consistency must give way to this plurality, in keeping with the complicated entirety of the integrated life. ‘The resulting ambivalence is part of the price we must pay to avoid the dehumanization of simple consistency in an unavoidably complex situation’, concludes Benjamin.⁴¹

In a similar vein, van Willigenburg suggests that compromising on some of our values in order to safeguard others might actually preserve (rather than undermine) the wholeness with which integrity is associated. But he does not think that ‘anything goes’: there must still be fidelity to one’s core principles, that is, to those that express one’s specific identity.42 And he also does not think that we can label someone a ‘person of integrity’ regardless of the particular principles to which they subscribe. Van Willigenburg refers here to integrity also involving purity: we should have no truck with principles that appear to be destructive of humanity, agency and personality as such. Van Willigenburg accordingly sees personal and moral integrity as part of the same continuum, according to which Hitler cannot be dubbed a man of integrity, and personal integrity necessarily involves a commitment to morality’s cause. The person of moral integrity could not therefore reach a principled compromise with Hitler. But we are still presented with a problem, since, generally, that person should stand up for what he or she believes are morality’s demands. On this account, morality is not a ‘negotiable commodity’ suitable for compromise.⁴³ It is here that the confusion objection arises: if we are considering compromise on moral matters then we must be confused about the nature of morality.

Numerous moral theories and theorists purport to tell us that they – and, apparently, they alone – can illuminate (or even dictate) the right way to act or the best way to be. To take but a few examples: deontologists point us towards the duties to be acted upon by moral agents, consequentialists steer us towards securing the best outcomes, while virtue ethicists describe the ideal character traits we should exhibit. What they all appear to share is an ‘objective’ outlook on good and bad, or right and wrong. We each occupy our own ‘moral space’, but if we encounter a ‘transgressor’, then morality should be able to distinguish which (if either of us) is right from which is wrong.⁴⁴ The ‘objective’ nature of morality, and hence the contended truth of a particular moral position, would appear to be immune from negotiation and barter.⁴⁵ And if we have reason to think that we are right (and therefore in the right) then, leaving aside extraneous considerations, why should we consider accommodating an opponent whom we judge to be morally mistaken?⁴⁶

These considerations lead some commentators to doubt that compromise is either conceivable or possible on those disputed moral issues, like euthanasia, which expose fundamental differences in moral commitment and outlook. Cohen-Almagor, for example, is sceptical about the prospects for compromise where ideology and identity are as intricately engaged as in a case like this,⁴⁷ and Tännsjö concludes similarly.⁴⁸ Yet, as we have seen, it is Holm who has particular difficulty on those occasions when:

First, I should clarify that I nowhere defend the view that the Netherlands exemplifies a compromise position on euthanasia (for which reason Holm does well to include his citation of my work before the semi-colon). Although one can detect a social compromise in the policy of pragmatic tolerance adopted in the Netherlands,⁵⁰ I see that jurisdiction as adopting a distinctively permissive approach to euthanasia, premised on such values as respect for autonomy and the obligation to eradicate suffering.⁵¹ I argue instead that a compromise approach would allow space for these values, whilst also affording room for competing (more prohibitive) arguments, premised on the intrinsic value of life and the alleged risks associated with allowing the practice.⁵²

However, secondly, and more importantly for present purposes, we must consider Holm’s resistance to compromise where one’s identity appears to be at stake. Like Golding before him,⁵³ Holm is open to the possibility of compromise where interests are contested, but he sees values as different, since these are more intimately connected with the disputants’ respective world views, and with all the metaphysical commitments these necessarily presuppose. Some ethical disputes will degenerate into irresolvable stand-offs, with the parties becoming more (rather than less) firmly entrenched and each pointing to the alleged absurdities entailed by the other’s claims. Although Holm does not mention it, there is an excellent illustration of this in the euthanasia literature, in the exchange between Harris (pro) and Finnis (anti) in Keown’s Euthanasia Examined.⁵⁴ Holm undoubtedly has disputants like these in mind when he mentions those to whom ‘the mere suggestion that there is a possible compromise is seen as grotesque and improper’.⁵⁵