Religious preferences in healthcare: A welfarist approach

Abstract

This paper offers a general approach to ethics before considering its implications for the question of how to respond to religious preferences in healthcare, especially those of patients and healthcare workers. The first section outlines the two main components of the approach: (1) demoralizing, that is, seeking to avoid moral terminology in the discussion of reasons for action; (2) welfarism, the view that our ultimate reasons are grounded solely in the well‐being of individuals. Section 2 elucidates the notion of religious preferences and describes the history and importance of their protection by human rights legislation. The following section defends the ‘Preference Principle’, according to which there is a reason to satisfy any preference (in so far as that satisfaction advances well‐being). Section 4 discusses the implications of this principle for religious preferences in healthcare, again seeking to bring out the special social and political importance of respect, and respect for such preferences in particular. The paper ends with a brief description of how to approach such problems from the perspective of a demoralized welfarism.

1. DEMORALIZING, WELFARISM AND HEALTHCARE

This may be an odd thing for a moral philosopher to say, but I think that morality is not fundamentally important.¹ In fact, I think it would be helpful if we stopped using, or at least drastically cut the use of, moral language in philosophical ethics, unless we are engaged in some nonnormative enterprise, such as describing a particular morality, that of common sense, for example, or of some particular group or individual. This is not because I am some kind of normative nihilist or rational egoist. I accept that we should do many things that morality requires us to do, such as not to inflict pointless suffering on nonhuman animals, but not that we should do them because morality says we should. Morality is a social phenomenon analogous to law, and in the case of law also I see no reason to do anything merely because the law requires it.

Another reason to avoid moral terminology in philosophical ethics is that morality functions through emotions, especially that of anger, of which the primary moral species is blame. The emotions, though they may have some cognitive content, are passions, and in most areas of philosophy, it is rightly thought that arguments should be assessed in the light not of emotion but of calm rational reflection. Blame is not entirely irrational, of course, but as Aristotle says, ‘it seems to listen to reason to some extent but to hear it incorrectly; it is like hasty servants who rush off before they have heard everything that is being asked of them and then fail to do it, and dogs that bark at a mere noise, before looking to see whether it is a friend. In the same way, spirit, because of its heated and hasty nature, does hear, but does not hear the command, and so rushes into taking revenge’.²

This is not to say that there could not be fundamental moral reasons. That is to say, morality could be more than a social phenomenon, constituting a set of independent norms, which must be characterized in moral terminology. (This picture of morality is analogous to the picture of the law in natural law theory, according to which positive law—the social phenomenon—can be assessed in the light of natural laws independent of positive law.) But we should not begin, as so many philosophers have done and continue to do, with the assumption that there are fundamental or ultimate reasons for action the content of which can be captured only by using moral terminology. We should introduce such reasons into our account only if they are independently justified and required to answer our ultimate practical question: what does one have reason to do? One can say, for example, that each of us has an ultimate reason not to inflict pointless suffering on a nonhuman animal without using any moral terminology. Someone might wish to add: ‘It is wrong to do so, and hence this reason is a moral one’. But since this introduces a whole set of moral notions and raises many questions about the nature and status of moral properties, the onus is on this person to explain the value of their suggested addition.

Moral language, then, including the notions of right and wrong, duty, rights, justice, virtues, and so on, is best avoided as far as possible in fundamental normative ethics. If someone claims that ϕ‐ing is wrong, for example, we should translate that as the claim that there is a reason, perhaps an overriding reason, not to ϕ, and then ask why.³ If the answer comes in moral terminology, then that will need to be translated as well. By demoralizing such language, we may arrive at what really matters—our reasons for action and what grounds them—and we will also be less likely to be misled by emotion.

What, then, does ground reasons? Nothing other, I suggest, than the welfare or well‐being of individual sentient beings.⁴ This is not a commitment to utilitarianism, since welfarism does not imply that the only grounding relation is that of impartial maximization, though I suggest that any plausible form of welfarism will allow that this is one way in which well‐being can ground a reason. But there may be others; it may be, for example, that we should give some priority to those who are badly off or that we should be especially concerned about the well‐being of those affected by our own agency.⁵ Nor are the only issues here purely ‘ethical’: matters involving, for example, the metaphysics of personhood or the theory of decision are bound also to arise.⁶

Demoralizing ethics strikes me as likely to be a particularly useful strategy to adopt when dealing with an issue such as that of the place of religion in healthcare. This area, especially the question of whether healthcare practitioners should be permitted to abstain from providing certain services on the basis of conscientious objection, is one where, even in philosophy, emotions run at a high level: there is plenty of what Julian Savulescu and Udo Schuklenk nicely describe as ‘breast‐beating’.⁷ There are also many appeals to moral rights, moral principles underlying liberal democracy, the moral duties of professionals, moral duties of toleration, moral integrity, and so on. There is little sign of increasing convergence or consensus, and that is not surprising, given the emotions in play and the fact that participants in the debate generally fail to get behind their principles to the reasons grounded in welfare that might support the implications of those principles in particular cases.

The case for welfarist demoralizing in the discussion of healthcare seems even stronger when we consider the primary aim of healthcare itself: to advance human well‐being through promoting health, whether or not we understand health as itself a constituent of well‐being. It is with that primary aim in mind that I suggest we approach the issue of religious pluralism in healthcare.

Note that my recommendation of demoralization is directed primarily at philosophers and others interested in working out the truth about how we should act, in the various circumstances we find ourselves in. I am not advocating total demoralization outside the study, in, for example, documents covering the behaviour of those working in health care. Morality and moral language clearly have great instrumental and facilitative values in certain contexts, as does positive law. There is a question, however, about how far philosophizing extends. I would recommend that those drawing up such codes, for example, seek to demoralize during the process. Finally, note that demoralizing does not justify dogmatism, inside or outside the study. I am recommending welfarism, and indeed demoralizing, in full awareness of the disagreement, there is about these and many other related issues. All I can do is describe the normative world as it appears to me.

2. RELIGIOUS PREFERENCES AND HUMAN RIGHTS

All individuals have preferences about how they themselves act and how others act in relation to them. Healthcare as a practice is constituted by acting and being acted on, sometimes in highly invasive and personal ways, and over the last half‐century or so, respect for the autonomy of individuals has become a central principle of biomedical ethics.⁸ This principle requires not only that treatment must be on the basis of informed consent, but that, in many cases, patients are offered some choice concerning the nature of that treatment. That choice may depend on the patient's attitudes to pain and discomfort, or risk, but also on preferences concerning the patient's life as a whole. An important class of the latter preferences is often described as ‘religious’, and they are usually related to the requirements of particular faiths. Examples might include a preference for the wearing of certain articles of clothing or religious symbols or for a certain diet or a ‘dispreference’⁹ for engaging in certain activities, such as the signing of a consent form on a holy day.

Note that I have spoken here of religious ‘preferences’ rather than beliefs. Unlike beliefs, preferences are not truth‐apt. If our discussion were to concern religious beliefs, it might be thought appropriate to require that those expressing them in a practical context be prepared to provide epistemic reasons for us to believe what they believe or at least to take what they believe seriously.¹⁰ But do we not describe preferences as reasonable or unreasonable? My preference that you stop torturing me seems reasonable, it might be said, while that for, say, this saucer of mud does not (unless I offer some reason for wanting it¹¹). This view of preferences, however, seems too intellectualist. Lots of our preferences are mere likes or tastes, and on the face of it, there is nothing unreasonable about liking something or having a taste for it, even a saucer of mud. This is not to say that I am understanding religious preferences to be mere tastes or recommending that we treat them as such. They are often bound up with a well‐worked‐out world view.¹² But, as I shall argue below, it is a mistake to think that for them to have practical import they must be provided with such a basis or indeed any basis.¹³ It may also be that advertising that decisions will be based on preferences, treated equally, rather than beliefs, epistemically assessed, will decrease the prevalence of any sense of one's, or one's religion's, having been disrespected.

Respect for the religious preferences of patients has emerged not only from the increasing weight attached to autonomy in medicine but also from the wider assertion of the right, the moral or human right, to religious freedom. Article 18 of the 1948 UN Declaration of Human Rights, for example, states:

Everyone has the right to freedom of thought, conscience and religion¹⁴; this right includes freedom to change his religion or belief, and freedom, either alone or in community with others and in public or private, to manifest his religion or belief in teaching, practice, worship and observance.

This right has been seen to extend to medical staff, and indeed medical institutions, as well as patients. Staff may also have preferences concerning symbols or food, for example, for proselytizing, or dispreferences for involvement in terminations, working on certain days, or exposing certain parts of the body. And such preferences and dispreferences can characterize collective entities, such as hospitals or local medical practices, independently of whether or not they are held by members working within them.

The fact that moral or human rights are ascribed to all has led to a growing focus on equality as their basis rather than liberty or freedom, and justifications offered for the legislation based on them usually refer to the prevention of discrimination rather than to interferences with liberty.¹⁵ But any such right is not, and cannot plausibly be seen as, absolute. If I have expressed a preference not to sign a consent form on Saturday for an urgent operation on Sunday, then, other things equal, it seems discriminatory not to make the form available to me on Friday. But if I insist on being treated only by fellow members of my own religion, or on obstructive and persistent proselytizing, or express a dispreference for being treated by those of a certain sexual orientation, any attempt to satisfy such preferences could itself be seen as discriminatory. The legislation, that is to say, rests on a defeasible principle, according to which religious preferences should be satisfied unless there is a good reason against doing so (I shall say more about such a principle in the following section). In the United Kingdom, any such nonsatisfaction must be proportional and a response to ‘a pressing social need’, such as public safety, health, or the protection of the rights and freedoms of others.¹⁶ This defeasible principle strikes me as reasonable, but as we saw above, a mere reference to human rights, whether seen as legal and/or moral, is not sufficient to provide an account of the normative reasons involved sufficient to explain which religious preferences are and which are not to be satisfied and why. This paper can be seen as a prolegomenon for such an account, as well as an illustration of the welfarist ethical method.

3. THE PREFERENCE PRINCIPLE AND WELL‐BEING

Let me return to demoralizing. I submit that basing respect for preferences in healthcare on a theory of significant human rights may be both unnecessary and insensitive. Consider a case in which a patient expresses a preference for one spoonful of sugar in their tea and, through inattention, is given two. It would be hard to argue that this patient's human rights have been violated. Equally, however, it seems that their preference should have been respected, and this might be used to argue that they had a right, if not a human right, to a single spoonful. But if we allow that the member of staff had a duty to respect their preference, postulating the right adds nothing to our understanding. It does identify the patient as the person to whom the duty is directed, but we already knew that. Duty can function alone, understood either as one duty among others or as overall. In fact, however, even the notion of duty may be unnecessary. Consider a case in which I can, if I wish, provide some small presents to a child. There are two presents, A and B, to choose from, and I know that the child has a very mild preference for A. But—again, through inattention—I give them B. Just as the notion of a human right was too weighty to illuminate the sugar case, so that of a duty here seems out of place. I had no duty to provide either present, and it will strike most people as excessive to claim that, once I have decided to provide a present, I have a duty to provide the best available, other things equal. But I certainly did have a reason to provide A rather than B. The notion of duty, then, can be set aside, as part of our demoralizing programme.

Consider the following demoralized and defeasible principle:

The Preference Principle: There is a reason to satisfy any preference.

On the face of it, this principle seems highly plausible. Because it is defeasible, and refers only to pro tanto reasons, it does not imply that there is an overall reason, for example, to satisfy sadistic or racist preferences (though it is not inconsistent with that claim). If we accept the Preference Principle, then, the question we might begin with is not why religious preferences should be satisfied, but why they should not. But once it is recognized that there may be a cost to satisfying such preferences, it can be seen that there is a need to understand the reasons for satisfying religious preferences in particular. As suggested above, reasons for action are grounded in well‐being. So we now need to answer the question of how satisfying preferences, religious or otherwise, might advance well‐being, and in particular—since healthcare is primarily ‘for’ patients—the well‐being of the patient. The answer one gives to this question will depend on one's theory of well‐being.

It is broadly agreed that something like Derek Parfit's tripartite categorization of theories of well‐being captures much of what is at stake between different accounts.¹⁷ Before I outline my own version of Parfit's triad, note a prior distinction, between substantive and explanatory accounts of well‐being. A substantive account concerns merely which items are good or bad for people. It might include, say, health, good interpersonal relations, and enjoyment, among many other things, including, of course, the satisfaction of preferences. An explanatory account goes deeper and addresses the question of why certain items are good or bad—in particular which properties of the items on the substantive list are ‘good‐making’ or ‘bad‐making’ for the individual in question. Fully to understand the connection between the preference principle and well‐being requires thought at the explanatory not merely the substantive level.

The three primary explanatory theories of well‐being are the following. First, hedonism, according to which what makes anything good is its promoting pleasure or enjoyment, and what makes anything bad is its promoting pain or suffering. Something will be overall good for a person in so far as it increases the overall balance of pleasantness over painfulness, most plausibly across their life as a whole.¹⁸ Second, desire theories, according to which what makes anything good is its fulfilling certain preferences, and what makes anything bad is its fulfilling a dispreference. What will be overall good for a person is what maximizes the net fulfilment of their preferences, most plausibly across their life as a whole. Finally, objective list theories suggest that what makes life good for a person is the presence in that life of certain items beyond mere pleasure and the satisfaction of desire, such as accomplishment, knowledge, or friendship. It might be thought that this kind of theory must be merely substantive, but that is not the case. The objective list theorist may claim that there are more fundamental, good‐making properties in goods such as accomplishment, such the property of fulfilling a person's human nature. But this is still an explanatory list theory since it includes an item independent of pleasure and desire‐satisfaction. According to a simpler theory, what makes accomplishment valuable just is its being a case of accomplishment (compare, for example, hedonism: ‘what makes a pleasure good just is its being pleasant’).

Let me now return to the Preference Principle and ask what each theory will suggest as the welfarist reason for satisfying a preference. The answer here is quite straightforward. According to hedonism, satisfying preferences is valuable in so far as it increases the patient's overall balance of pleasure over pain. Imagine a case in which I am asked to choose between two items. I judge them to be equal but plump for A over B. I do have a genuine preference for A, but if I really do not care which item I receive (it is just a whim, perhaps for a saucer of mud), it may be that on hedonic grounds there is no reason to satisfy my preference. On a standard desire account, however, my receiving A will benefit me more than my receiving B, because it satisfies a preference. Objective list accounts are in this respect like hedonism and will not see any value in the mere satisfaction of desire in itself.

Which theory of well‐being is correct? I want first to suggest that we put desire theories to one side. Consider my whim for the saucer of mud. The Preference Principle implies that there is a defeasible reason to satisfy that preference. But if I get nothing from it, and would not experience any frustration were the preference to remain unsatisfied, it is hard to see how in this case the principle is applicable. Parfit himself made this point well with the following example:

I tell you that I am about to make your life go better. I shall inject you with an addictive drug. From now on, you will wake each morning with an extremely strong desire to have another injection of this drug. Having this desire will be in itself neither pleasant nor painful, but if the desire is not fulfilled within an hour it will then become very painful. This is no cause for concern, since I shall give you ample supplies of this drug. Every morning, you will be able at once to fulfil this desire. The injection, and its after‐effects, would also be neither pleasant nor painful. You will spend the rest of your days as you do now.¹⁹

Hedonism provides a more plausible account of the reason to satisfy a preference. Often, satisfying a desire will produce pleasure and avoid pain. Consider the simple case of a spoonful of sugar. Even if the patient found the tea overall pleasant, it would have been more pleasant had their preference been satisfied. And they also may suffer at least an unpleasant twinge of annoyance that their expressed preference has been ignored. Any plausible objective list account will include pleasure on its list. (Hedonism could be described as a nonobjective list account, with one item on its list.) But objectivists can argue that the satisfaction of preferences can also play a part in constituting nonhedonic goods. Consider, for example, the value of interpersonal relationships. The religious practice provides many people with an opportunity for interpersonal relationships, which are often deep and long‐lasting. The satisfaction of an expressed religious preference within a healthcare context can be seen as part of what it is to participate in a religious community and benefit from the relationships it makes possible. Many believe that religion also gives meaning to their lives, and living in accordance with a religion might also be said to be an accomplishment, or to advance the observer's understanding. (A hedonist, remember, can also refer to these goods but only substantively, as sources of—often deep—enjoyment and contentment.)

4. RESPECT AND THE SIGNIFICANCE OF RELIGIOUS PREFERENCES

The fact that religious preferences tend to matter so much to their possessors is one reason why they are singled out as especially weighty in healthcare practice, policy, and management. But as we saw above this is not the only reason. Satisfying religious preferences is also seen as required by principles of equal respect, which underlie, for example, article 18 of the UDHR. Can such respect be incorporated into the welfarist account I have sketched above? A hedonist cannot of course postulate respect as a value in itself. But it is all too obvious how much distress has been, and is, caused to members of various, often oppressed or marginalized, groups through disrespect. Respectful treatment of others can also provide longer‐term benefits by improving the general status of marginalized groups within society, and hence, as the classical hedonistic utilitarians knew only too well, increasing general happiness. In addition to these points, the objectivist might point to the damage lack of respect can cause in other domains of welfarist value, such as loss of accomplishment or damage to personal relationships through lack of esteem or self‐esteem. Could lack of respect be said to be a harm in itself? This is a harder claim to defend, once the effects of disrespect on these other values have been taken into account. Merely being disrespected by some other person, if it has no deleterious hedonic effects on my consciousness or on other aspects of my well‐being, seems harmless, though of course, we may wish to criticize the disrespecting individual for doing something that is generally harmful, often seriously so.

This brings us to the question of why religious preferences in particular are given such prominence in health care. The answer lies primarily in human history. For thousands of years, at least since the invention of agriculture, groups of human beings have engaged in hostile, often violent, conflict with other groups. Membership of any group has depended on a variety of characteristics, some, such as spatial or genetic origin, having no relation to religion or other cultural practice. But such practices have nearly always been central to the social identity of members of such groups, who have often been identified partly through their engagement in such practices and indeed persecuted for them (sometimes because such persecution is itself required by the norms constituting some other religion or belief‐system). Whether a religious belief is or is not rational is an interesting question, but it is not relevant to the issue of whether religious preferences should be given weight in healthcare. They are to be given such weight because failure to do so is not only to risk harm to the disrespected individuals but, through the resentment at that very disrespect, to risk perpetuating the harmful and unproductive—and largely irrational—conflicts between groups of human beings that continue to characterize our species.

As we have seen, other things being equal, there is a reason to satisfy any preference. Some preferences matter significantly more to their possessors than others, and religious preferences tend to be within this group. That, along with the role they have played and continue to play in human history, justifies giving them special weight in healthcare and not merely for health‐related reasons. These welfarist and historical factors also explain why religious preferences, which in themselves express hostility to members of other groups, should in general not be satisfied, whether or not those groups are themselves religious. And it is not only hostile religious preferences that should often remain unsatisfied. Consider some religion according to which viruses have high moral status. The appeal of a doctor to freedom of religion to justify their not distributing a vaccine during a pandemic should clearly be overruled on the grounds of harm to patients.²⁰ The arguments for respecting preferences based on, say, sexual orientation or nonreligious world views, such as pacificism or vegetarianism, are similar. They can matter greatly to their possessors and have often been the target of hostile and disrespectful treatment by others. But of course, this is not to say that preferences of a kind that has not been targeted by others can be ignored, especially if that preference matters a lot to its possessor. Even an unimportant preference—such as that for a single spoonful of sugar—should not be ignored without reason.

I have been focusing largely on the religious preferences of patients, but the same arguments carry across to medical staff and institutions. These preferences matter greatly to the individual staff, including those whose activity constitutes that of the institution in question, and deserve particular respect because of their historical role. Equally, however, they can never be taken as overriding. They may be in conflict, perhaps disrespectful conflict, with the preferences of others, religious or nonreligious, or with reasonable, welfare‐promoting requirements imposed by healthcare authorities.

It might be argued that religious preferences in such cases are themselves inconsistent with the aims of a healthcare system, in such a way that the religious preferences of medical staff should never be taken into account. As Julian Savulescu notes, ‘The primary goal of a health service is to protect the health of its recipients’.²¹ But this is not to say that it does not have secondary or further goals, one of these being the well‐being of staff. It may even be that in some cases (e.g., where the cost to the patient is negligible, such as the painless delay of a procedure by a very short period of time) staff religious preferences can outweigh the preferences or health of patients. But have medical practitioners not promised to put the health of their patients first?²² Yes, but like any promise, this one is defeasible. If a doctor is about to offer a patient a mild pain‐killer for their headache, and then notices that they can save the life of a stranger instead (perhaps without using their medical skills), we would expect them to do the latter.

When such cases of conflict are likely not to be unusual, a policy decision has to be taken by some higher authority, and it may well be that certain religious preferences are insufficiently weighty to defeat competing considerations. This is particularly true of institutional preferences the overriding of which need not involve existing members of the medical staff in activities for which they have a strong religious dispreference. Consider, say, a religious employer which has a preference that its employees not be offered contraceptive advice funded by its healthcare plan and objects to any government requirement that such advice is made available.²³ That religious preference should, other things equal, be satisfied. But the effect on current employees of its being satisfied could be devastating, while—however ‘strongly held’ the preference is—it is highly implausible that the individual well‐being of those who constitute ‘the employer’ (the senior management team, perhaps) is going to be seriously harmed to any great extent: the worst they will experience is moderate anger for a relatively short period. Further, the availability of such advice could plausibly be said to be part of the movement toward the emancipation of women through increasing reproductive freedom, and this provides a contextual historical argument against the satisfaction of the employer's preference analogous to that noted above for religious preferences.²⁴

I have dealt with this final case somewhat briefly, but I hope that the discussion brings out how the welfarist structure within which I have placed this particular issue can be used to resolve conflicts arising from religious preferences without appealing to any moral notions, such as a right to religious freedom, a right to reproductive freedom, fairness, or justice. These and similar notions are often too vague to be helpful, can distract us from what ultimately matters in such conflicts—the well‐being of those involved—and may encourage us in following emotion rather than reason. Faced with the question of how to confront any conflict of preference, religious or otherwise, there are three stages to a solution: demoralize; assess how the well‐being of those involved will be affected by various salient outcomes²⁵; and make a judgement about the best outcome on the basis of that assessment alone.

CONFLICT OF INTEREST

The author declares no conflict of interest.

Biography

Roger Crisp is director of the Oxford Uehiro Centre for Practical Ethics, professor of moral philosophy at the University of Oxford, and Uehiro fellow and tutor in philosophy at St Anne's College, Oxford. He has published many articles on ethics, including one in 1987 on euthanasia in the first number of Bioethics. His most recent book is Sacrifice regained: Morality and self‐interest in British moral philosophy from Hobbes to Bentham (Clarendon Press, 2019).

Crisp, R. (2023). Religious preferences in healthcare: A welfarist approach. Bioethics, 37, 5–11. 10.1111/bioe.13114