Ethical Theories and Controversial Intuitions

ABSTRACT

We have controversial intuitions about the rightness of retributive punishment, keeping promises for its own sake, and pushing the heavy man off of the bridge in the footbridge trolley case. How do these intuitions relate to ethical theories? Should ethical theories aim to fit with and explain them? Or are only uncontroversial intuitions relevant to explanatory ethical theorising? I argue against several views that we might hold about the relationship between controversial intuitions and ethical theories. I then propose and defend the view that ethical theories should only aim to fit with and explain the intuitions that almost all people (who minimally understand the relevant issues) share. I argue that this view has interesting and important implications for ethical theorising and theorising about justice and equality.

KEYWORDS:

• Moral disagreement
• normative ethics
• ethical theories
• intuitions
• moral methodology

1. Introduction

We have controversial moral and ethical intuitions. Some have the intuition that it is right that wrongdoers are punished even when there are no good consequences (rehabilitation or deterrence) to their punishment. Others don’t have this intuition.¹ Some have the intuition that there is strong moral reason for us to keep our promises even when there are no good consequences brought about by our doing so.² Others have the intuition that this is not the case.³ Some have the intuition that it’s wrong to push the heavy man off of the bridge in the footbridge trolley case; others have the intuition that it’s right to push the heavy man.⁴

This paper is not concerned with the question of whether controversial moral intuitions can justify (controversial) moral beliefs.⁵ Rather this paper is concerned with the question of whether controversial intuitions have any legitimate role in explanatory normative ethical theorising. Part of what normative ethical theories aim to do is to explain why we ought to do what we ought to do. Contractualism, rule-consequentialism, utilitarianism, virtue ethics, Rossian pluralism, and Kantian views are all normative ethical theories with this explanatory aim. For instance, contractualist and rule consequentialist views aim to provide a deep unifying explanation of why we ought to perform and refrain from performing a diverse range of actions (refrain from harming others, help others, keep our promises, etc).⁶

We should contrast the project of partially explanatory normative ethical theorising with the project of making first-order arguments in applied ethics. Some arguments in applied ethics do not aim to explain why we ought to do what we ought to do but rather only aim to make a case for a particular view about what we ought to do. For instance many, arguments for the view that abortion is permissible or that eating meat is wrong aren’t aimed at explaining why we ought to do what we ought to do but rather at showing that we ought to do/not do certain things. So, we should distinguish:

1. whether controversial ethical intuitions have a legitimate role in explanatory normative ethical theorising;

2. whether controversial ethical intuitions can justify (controversial) ethical beliefs; and

3. whether controversial ethical intuitions play a legitimate role in all arguments in applied ethics.

In this paper I am only concerned with (a) and not (b) or (c). So, everything that I say in this paper is compatible with the view that controversial intuitions can and should play a role in arguments in applied ethics.⁷

There has been very little explicit discussion of whether explanatory normative ethical theories should fit with and explain controversial intuitions. In this paper I investigate how normative ethical theories relate to controversial intuitions. I discuss several principles that state views about the relationship between controversial intuitions and ethical theories. Almost everyone believes that intuitions play a role in supporting ethical theories. The principles that I discuss aim to give the intuition-related desideratum on ethical theories. That is, they aim to provide a principled view about how a theory is better as a normative ethical theory because it fits with our intuitions (controversial or otherwise). They aim to explain how intuitions, and which intuitions, support normative ethical theories. There will be other desiderata for an ethical theory, that is, other features of a theory that make it better as a normative ethical theory, and which give us reason to accept it. For instance, normative ethical theories will plausibly also be better to the extent that they are simple, elegant, action-guiding, determinate, and/or coherent.⁸ This paper focuses on just one desideratum: the desideratum regarding how it is a virtue of an explanatory normative ethical theory that it fits with and explains certain intuitions.

In §2 I consider a relativist view according to which it is a desideratum of a normative ethical theory that it fits with and/or explains the ethical intuitions of the person considering the theory or doing the theorizing. An alternative to this relativist view is:

Widely Shared. It is a desideratum of an explanatory normative ethical theory that it fits with and/or explains ethical intuitions that are widely shared.⁹

In §3 I argue that we should reject several ways of understanding or specifying Widely Shared. In §4 I argue that we should accept the following specification of Widely Shared:

Shared by Almost All. It is a desideratum of an explanatory normative ethical theory that it fits with and/or explains the ethical intuitions that almost all people who minimally understand the relevant issue(s) share.

I argue that there are advantages to accepting Shared by Almost All, that it avoids all the problems that face the relativist view and alternative specifications of Widely Shared, and I address objections to Shared by Almost All. In §5 I argue that Shared by Almost All has interesting and important implications for particular intuitions that are sometimes used in ethical theorizing, for the methodology of ethical theory, as well as for theorising about justice and equality. In the rest of this introduction I’ll clarify some background assumptions of this paper.

The question that I’m concerned with in this paper is whether controversial ethical intuitions can play a legitimate role in a particular type of social/public philosophical theorising – explanatory normative ethical theorising – rather than whether such intuitions can play a legitimate role in justifying individual ethical beliefs. So, the legitimate role of intuitions in a particular practice, our practice of engaging in public explanatory normative ethical theorising, is the object of inquiry. The view that I argue for in this paper, Shared by Almost All, gives controversial intuitions very little role in explanatory normative ethical theorising. In order to avoid self-defeat, the arguments that I make for Shared by Almost All, and against competitors to it, should not be justified on the basis of my own controversial ethical intuitions. Instead I often utilize considerations regarding our current practice of normative ethical theorising in my arguments. I am investigating the extent to which controversial intuitions can play a legitimate role in our social practice of public explanatory normative ethical theorising. An account that strays too far away from our current practice would not be an account of our practice of public explanatory normative ethical theorising but of something else. So, I aim to give an account that provides a principled and plausible view of the extent to which controversial intuitions can play a legitimate role in public explanatory normative ethical theorising that, as far as possible, also fits with our current practice of such theorising.¹⁰

I make three further background assumptions. I assume that intuitions are intellectual seemings and are what is being reported when moral philosophers consider a case and give a verdict about it.¹¹ I assume that when people respond to surveys about moral issues and give their verdicts about moral cases they are reporting their moral intuitions.¹² And I assume that whether an intuition is an intuition about a principle or a case does not in itself matter for whether it can be legitimately utilized in normative ethical theorising.

2. Relativism

In this section I’ll focus on views according to which whether an intuition is controversial or not makes no difference to its role in supporting an explanatory normative ethical theory. One such view is a non-starter: namely the view that it is a desideratum of a normative ethical theory that it fits with and explains intuitions that track the truth. This view holds that whether an intuition is controversial or not doesn’t matter for the purpose of normative ethical theorising. But intuitions are redundant in this desideratum: we may as well hold that it is a desideratum of a normative ethical theory that it tracks the ethical truth. This shows that this candidate desideratum cannot be the intuition-related desideratum on a normative ethical theory.

Alternatively, consider

Relativism. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains the person considering the theory or doing the theorising’s ethical intuitions.

According to Relativism, our controversial intuitions are just as important as our non-controversial intuitions for the purpose of explanatory ethical theorising. I call this view Relativism because it holds that what counts as a good ethical theory is, to a certain extent at least, relative to the theorist or person assessing the theory’s intuitions. So, on this view it’s not the case that there are better and worse ethical theories in terms of how they meet the intuition-related desideratum on ethical theories and we, with our particular intuitions, can be attracted to ethical theories that don’t happen to be the best. Rather, what counts as a better or worse ethical theory partially depends on, and so is partially relative to, our intuitions in particular.

There are at least four problems with Relativism. The first problem is that it clashes with our practice of explanatory normative ethical theorising. Even Frances Kamm – who many believe relies on obscure or baroque intuitions – is at pains to stress that she does not take the ethical theory that she constructs to be one that just fits her particular intuitions; rather she thinks that it is important that others are likely to come to have intuitions that accord with hers if they soberly and seriously consider the cases that she discusses and the distinctions that she makes.¹³ And, standardly, ethical theorists claim that it is ‘widely shared’ intuitions that play a justificatory role in ethical theorizing rather than just the intuitions (no matter how controversial) of the theorist.¹⁴

Second, Relativism cannot explain why intuitions about the morality of abortion and the moral status of animals do not feature in the construction and justification of explanatory normative ethical theories. And why considerations regarding these controversial issues play next to no role in arguments for or against ethical theories such as virtue ethics, contractualism, or rule consequentialism. If Relativism holds, then these intuitions should play as much of a role in normative ethical theories and theorising as uncontroversial ethical intuitions.

It might seem that controversial intuitions about the moral status of animals sometimes feature in objections to contractualism as an explanatory ethical theory. For several ethicists have argued that contractualism generates the wrong results regarding cruelty to animals. So, controversial intuitions do feature in our practice of assessing explanatory ethical theories. However, those who make this argument do not take themselves to be – and plausibly are not – relying on controversial intuitions. For instance, Alastair Norcross objects to contractualism on the basis of its treatment of animals. But he claims that contractualism does not account for ‘our common-sense attitudes towards animals’. This is because ‘[t]he prevailing view may be that animals’ interests are not as significant as those of humans, but it is not that they count for nothing’. But according to Norcross, contractualism entails that animals’ interests count for nothing.¹⁵

Third, Relativism conflicts with the widely held view that (ethical) intuitions relate to ethical theories as (scientific) observations relate to scientific theories.¹⁶ Scientific theories don’t aim to just explain the observations of the scientists who propose them or individuals who are assessing them; they aim to explain the observations of all scientists and experimenters who conduct similar experiments to the observations that these theories are designed to explain. Suppose that theory T is motivated by the fact that it explains observation O1 in experimental circumstances C that some scientists or other experimenters have made. But other scientists conduct further experiments in similar experimental circumstances to C and do not make similar observations. In this case either T must explain why they did not make similar observations or the fact that T explains observation O1—which some make but others do not – no longer motivates T.¹⁷ But if it is only the intuitions of the theorist or person considering the ethical theory that matter for whether that ethical theory is a good one or not, then intuitions do not relate to ethical theories as observations relate to scientific theories.¹⁸

Should we think of intuitions as relating to ethical theories in the way that observations relate to scientific theories? There is a good prima facie case that we should think of intuitions as playing this role if they are to play any role in ethical theorising. As I explained in §1, the ethical theories that I am concerned with are at least partially explanatory theories. Scientific theories share a similar explanatory aim. This gives us reason to think that the starting points of both explanatory scientific theories – observations—and explanatory ethical theories – intuitions—should be treated similarly. Furthermore, the fact that common observations can be made by anyone with the relevant equipment and that scientific theories are based on such common observations separates scientific theories from pseudo-scientific theories, and to this extent gives us reason to take scientific theories seriously.¹⁹ Similarly if ethical theories are based on common intuitions rather than the intuitions of the theorist alone, this gives us and others reason to take them seriously.

Fourth, it seems to many that the moral theory that we should accept should be a theory that we can justifiably ask that others adhere to as well as that we should ourselves adhere to. For morality is an interpersonal endeavour that is plausibly centrally about whether our performing certain actions is justifiable to others. But if we accept Relativism, then the best moral theory will not necessarily be one that we can justifiably ask anyone else to adhere to. For it does not seem that we can justifiably ask others to adhere to the demands of a theory that is partially justified on the basis that it fits with intuitions that they and many others do not share.

It might seem that this fourth argument against Relativism relies on the truth of a form of contractualism. But this is not the case. For act-consequentialists and rule-consequentialists both endorse the view that our ethical theories should be justifiable to everyone.²⁰ They just hold that their theories are justifiable to everyone because they take the well-being of all into account equally.²¹

So, there are four reasons to reject

Relativism. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains the person considering the theory or doing the theorising’s ethical intuitions.

3. Philosophers’ Intuitions and Majority Intuitions

Brad Hooker and Mark Timmons amongst others claim that

Widely Shared. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains ethical intuitions that are widely shared.²²

Widely Shared doesn’t face the problems that Relativism faces. And, given that we have reason to reject Relativism, we’ll need to accept some version of Widely Shared in order to accommodate the view that intuitions play some important positive role in ethical theorising.

Widely Shared is quite vague, and there are good reasons to reject certain natural ways of understanding it. First, we might understand Widely Shared as involving the following view:

Shared by the Majority of Philosophers. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains the ethical intuitions that the majority of moral philosophers share.

However, we would only be justified in holding this view if we were justified in holding that the intuitions of moral philosophers are more reliable than the intuitions of non-moral-philosophers. But many contest this. First, many moral philosophers and metaethicists argue that we do not (and should not) treat others as experts on moral matters; many argue that we should not defer to the purely moral judgments of others – as opposed to the partially empirical judgments of others – or that we should not defer to the moral judgments of moral philosophers or treat them as more reliable on moral matters than non-moral philosophers.²³ It is widely recognized in metaethics that moral realism faces a prima facie challenge to explain why such deference, and the identifiable expertise it implies, is dubious in the moral case.²⁴ In this case, normative ethicists should not rely on the view that moral philosophers should be deferred to by non-moral-philosophers. For the view that there are identifiable moral experts is a view that is too controversial for a central methodological principle in normative ethical theorising to rely on. But if moral philosophers’ intuitions about ethical matters were in better epistemic standing than the intuitions of non-moral-philosophers, then non-moral-philosophers ought to treat moral philosophers as experts on ethical issues and defer to them on these topics.²⁵

Second, many virtue ethicists claim that moral philosophers’ intuitions are not more reliable than the intuitions of virtuous non-philosophers.²⁶ So, Shared by the Majority of Philosophers won’t be acceptable to virtue ethicists, and may entail the falsity of virtue ethics. And, other things equal, we should reject an account of the intuition-based desideratum on ethical theories that cannot be accepted by proponents of all schools of ethical theorising. For such accounts will fit poorly with our practice of explanatory normative ethical theorising (see §1).²⁷

Alternatively, we might understand Widely Shared in the following way:

Shared by the Majority. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains the ethical intuitions that the majority of people share.

The problem with Shared by the Majority is that it makes the intuitions that it is desirable that ethical theories fit with too contingent on the majority intuitions at a particular time or over time.

There are many intuitions that the majority of people hold that ethical theories are not taken to need to fit with and/or explain. The majority of people in the US believe the following:

1. the death penalty is morally acceptable;

2. polyamorous relationships are morally wrong; and

3. animal cloning is wrong. ²⁸

A majority of people also believe that the following considerations are relevant to whether an action is right or wrong: whether an action involves

• (4) disrespect for an authority figure;

• (5) disrespect for the traditions of their society; or

• (6) doing something disgusting, impure, ’unnatural’, or degrading.²⁹

Normative ethical theories and theorists do discuss a plethora of intuitions but they do not discuss intuitions (1–3) or the intuitions that whether an action involves (4–6) is relevant to whether it is right or wrong. And explanatory normative ethical theories do not seem to fit with and/or explain these intuitions. Kantian, Rossian, and act-consequentialist views, for instance, seem to entail that whether an action involves (4–6) is not morally relevant. None of these considerations appear in Ross or Rossians’ lists of prima facie duties. Never acting on the maxim ‘never do what is traditional’ is perfectly universalizable, and acting in this way does not involve treating anyone as a mere means. But, if we can, we should accept a version of the intuition-based desideratum for normative ethical theories that fits with our practice of ethical theorising and does not entail that most widely held and discussed contemporary ethical theories do poorly on that desideratum (see §1 above). So, we should not accept Shared by the Majority.

It might be objected that although a majority of Americans believe (1–3) and that whether an action involves (4–6) is relevant to whether it is right or wrong this is because they believe that there are deterrence-based good consequences to the death penalty; they believe that polyamorous relationships have negative consequences for those involved in them, their children, and society at large; and they believe that disrespecting authority figures and traditions has bad consequences for those who disrespect them (and their friends and/or family) as well as society more generally. All ethical theories can accommodate these judgments. And so this data does not present a problem for Shared by the Majority. Call this objection to my argument against Shared by the Majority the consequences objection. In the rest of this section I’ll give three reasons why the consequences objection fails.

First, it is hard to see how the 65% of Americans who believe that animal cloning is wrong could all believe that animal cloning would have sufficiently bad consequences to make it wrong.³⁰ And there is empirical evidence showing that those who believe that doing something disgusting, impure, ‘unnatural’, or degrading is relevant to whether that action is wrong do not believe this because they believe that there are bad consequences to doing disgusting, impure, or degrading things.³¹ I recruited 105 participants from Amazon’s Mechanical Turk Platform. I gave them two vignettes in which someone trivially disobeyed an authority figure (in one case their boss, in another case their parents) without anyone finding out and where their doing so was (realistically) stipulated to have no bad consequences at all. (In the former case they disobeyed by failing to do up their button or wear a tie; in the latter case they just broke an arbitrary curfew). I asked participants whether they agreed with the statement that, putting aside any good or bad consequences of their disobeying the authority figure, it was wrong for those individuals to so disobey. In one case 71% of participants agreed with this statement in the other case 55% of participants agreed with this statement.

Second, even if the consequences objection succeeded, this would not detract from what is, in a sense, the main point of my argument against Shared by the Majority. Even if the majority’s intuitions that (1–3) hold are due only to the good consequences they perceive of the death penalty and the bad consequences they perceive of polyamory and animal cloning, it could easily have been the case that the majority thought these things were right/wrong for their own sake. Similarly, suppose that the majority only think that whether an action involves (4–6) is morally relevant because they believe that there are bad consequences of such disrespect and doing disgusting and degrading things. Still the majority could easily have believed that it is morally relevant whether an action involves (4–6) because such disrespect and disgusting/degrading conduct is morally bad in itself. For some people clearly believe these things.³² Shared by the Majority makes an important desideratum of an ethical theory hostage to the tyranny of the contingent intuitions of the majority. What the intuitions of the majority currently are is not relevant to the moral truth. But if Shared by the Majority held, the current intuitions of the majority or the majority over time would be relevant to the moral truth. In the rest of this paper I’ll try to articulate a view of the relationship between controversial intuitions and ethical theories that avoids making ethical theories subject to the contingent and morally arbitrary tyranny of the intuitions of the majority but also avoids the problems that relativism faces.³³

4. Intuitions That are Shared by Almost All

In the last section I raised problems for adopting several versions or specifications of

Widely Shared. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains ethical intuitions that are widely shared.

I propose that we accept the following specification or interpretation of Widely Shared:

Shared By Almost All. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains the ethical intuitions that almost all people who minimally understand the relevant issue(s) share.

Where to minimally understand the relevant issues is just to in fact grasp the issues involved. For instance, to minimally understand a complex trolley case is to accurately represent all of its non-moral features and to have a minimal competence with the moral concepts involved.

Shared by Almost All avoids the problems that the versions of Widely Shared that I discussed in §3 face. It does not entail that moral philosophers or anyone else are moral experts. It does not entail that it is a virtue of an ethical theory that it fits with and explains people’s intuitions that we should do as authority figures demand, as the traditions of our societies require, or that we shouldn’t do things that render us impure.

Shared by Almost All also has significant virtues. Shared by Almost All fits with the view, which I argued that we have some reason to accept in §2, that intuitions relate to ethical theories as observations relate to scientific theories. It explains why controversial intuitions about abortion and euthanasia play no role in arguments for ethical theories such as rule consequentialism, contractualism, and Rossian pluralism. It entails that the ethical theory that best satisfies the intuition- related desideratum will be to that extent justifiable to almost everyone because it fits with and is based on their intuitions. And it fits with our first-order practice, exhibited by Kamm, Hooker, and Timmons, of holding that ethical theories need to fit with and/or explain widely shared intuitions. Shared by Almost All accordingly avoids the problems that I argued that relativism faces in §2. In the rest of this section I’ll clarify Shared by Almost All and address some objections to it. I’ll do this by answering a set of questions that it is natural to ask about it.

Question:

Which intuitions are shared by almost all?

The intuitions that are shared by almost all will include intuitions such as:

1. It is sometimes wrong to kill others;

2. We should sometimes help people; and

3. We should sometimes keep our agreements.

People are inclined to accept a variety of different justifications for these intuitions: prudential/self-interested, consequentialist, rights-based, virtue-based, etc. But it seems that almost all people share the intuitions that (a-c) hold. I recruited 196 participants from Amazon’s Mechanical Turk Platform and asked them whether they agreed with these statements. 99% agreed with (a); 97% agreed with (b); and 99% agreed with (c). This provides us with prima facie evidence that almost all people agree with these claims.

Philippa Foot claims that it could not be the case that the ultimate moral principle is that it is wrong to walk around a tree clockwise or that it is wrong to look at hedgehogs in the light of the moon.³⁴ I suspect that almost all people who have considered and understood these claims will agree with Foot. Michael Huemer gives several paradigm examples of ethical intuitions including the following:

Enjoyment is better than suffering;

If A is better than B and B is better than C, then A is better than C;

It is unjust to punish a person for a crime that he did not commit; and

If a person has a right to do something, then no person has a right to forcibly prevent him from doing that thing.

He says of these propositions that ‘prior to entertaining arguments for or against them, each of these propositions seems true’.³⁵ I suspect that almost all people share Huemer’s intuitions that these propositions are true.³⁶

It has been argued that we have good reason to believe that almost all people who consider and understand (the right version of) the experience machine thought experiment have the relevant anti-life-in-the-experience-machine intuition. In a survey using (the right version of) the experience machine thought experiment it was found that over 90% of participants had the anti-life-in-the-experience-machine intuition³⁷ So, the anti-life-in-the-experience-machine intuition is also plausibly shared by almost all. ³⁸

Question:

Does Shared by Almost All entail that moral philosophers have no more moral expertise than non-philosophers and non-moral-philosophers?

No. In the literature on philosophical expertise it is common to distinguish between two hypotheses:

1. Philosophers have better intuitions than non-philosophers; and

2. Philosophers make better use of intuitions in theory construction than non-philosophers.

Defenders of philosophical expertise now often hold that (2) does not entail (1) and defend (2) rather than (1).³⁹ And Shared by Almost All is consistent with the view that philosophers can make better use of their intuitions in theory construction. Furthermore, Shared by Almost All does not entail that (1) is false. It just entails that for the purpose of explanatory ethical theorising it is a virtue of an ethical theory that it fits ethicists or philosophers’ intuitions only if these intuitions are shared by almost all – for the kinds of reasons discussed in §2.

Question:

What exactly does ‘almost all’ mean in Shared by Almost All?

I propose that

Almost All. For the purpose of Shared by Almost All an intuition counts as shared by almost all people iff (i) it is shared by all people or (ii) it is shared by a substantial majority of people but we have good reason to accept an error theory of the intuitions or lack thereof of the people who do not share the relevant intuition.

I’ll give some examples of when we may have good reason to accept such an error theory. We plausibly have good reason to believe that psychopaths don’t have moral concepts because their motivational states aren’t linked up to their putative moral concepts in the right way: they are not motivated to refrain from performing actions that they judge to be wrong for instance. According to some this – combined with the fact that they find it hard to make the moral/conventional distinction – shows that they don’t really have moral concepts.⁴⁰ So, if an intuition is shared by all people except psychopaths, then this does not stop that intuition from being one that ethical theories should fit with and/or explain. This is because we have a good error theory of these people’s lack of the relevant intuition, namely that they don’t genuinely have moral concepts.

It might seem that if we understand Shared by Almost All in this way, then it will entail that the intuitions that I above claimed to be shared by almost all – it’s sometimes wrong to kill others, we should sometimes help people, and we should sometimes keep our agreements – are not in fact shared by almost all because moral skeptics do not share these intuitions.⁴¹ However, moral skeptics will have had these intuitions prior to adopting moral skepticism. And even after adopting their skepticism, skeptics generally attempt to explain the correctness of these intuitions: they hold that these intuitions track normative truths because doing (refraining from doing) these things serves our desires or interests. (What of error theorists? Most error theorists share the moral intuitions of others but merely think that there are good arguments for holding that, ultimately, their intuitions are off-track).⁴²

I have encountered a different objection to Almost All. It has been put to me that the combination of Shared by Almost All and Almost All entails that if 100% of people – who do not get disqualified by (i) or (ii) above – have an intuition, then an ethical theory must fit with and/or explain that intuition. But if only 99% of people share that intuition, then an ethical theory need not fit with and/or explain this intuition. But this seems arbitrary. It seems that we should not accept a view that entails that there is such a harsh cut-off point regarding the percentage of people who share an intuition and whether a theory should explain that intuition.

However, Almost All does not entail that there is such an arbitrary harsh cut-off point. For if 99% of people have an intuition, and 1% do not, then we have good reason to believe that either the 1% of people have not genuinely understood or reflected on the relevant issues or that there is a good error theory of their intuitions such as that they are mistaking one concept for another relevant concept, they don’t genuinely have moral concepts, or they are being misled by an evolutionary (or other irrelevant) influence or bias. For otherwise it would be hard to explain why the 99% have a moral intuition that the 1% do not have. If the split is 98%/2% we also have reason to believe that there is an error theory, lack of understanding, or malfunction regarding the 2%’s intuitions but this reason is weaker (and so on).⁴³ But if there is a substantial minority of people who have different intuitions from those of the majority, we do not obviously have any reason to believe that there is some lack of understanding, error, or malfunction with the significant minority.

Question:

In §3 it was argued that Shared by the Majority is too contingent on the morally arbitrary intuitions of the majority. Isn’t Shared by Almost All just as contingent on an equally morally arbitrary fact?

I don’t believe so. Which intuitions are shared by just the bare majority of people fluctuates quite a lot over time. For instance, in the last 10 years the majority has shifted from thinking that same-sex marriage is wrong to thinking that it is permissible.⁴⁴ And similarly, the majority intuitions about the death penalty, abortion, and euthanasia seem to shift relatively quickly with the political climate.⁴⁵ In contrast, the intuitions that are shared by almost all are a relatively stable set. Remember that intuitions that are shared by almost all are intuitions such as that it is sometimes wrong to kill others, we should sometimes help people, and we should sometimes keep our agreements. And whether almost all people who understand the relevant concepts have these intuitions does not seem to change over time.

Anthropologists, psychologists, and sociologists have developed a variety of different views that entail that these intuitions will not change over time and explain why this is so. I’ll take one example. The Human Relations Area Files is an archive of thousands of full text ethnographies spanning several hundreds of years from hundreds of different societies of all different types from all different areas of the globe. Some anthropologists have argued that an analysis of this archive shows that all moral systems that have existed are devices to facilitate cooperation.⁴⁶ This view predicts that almost all people – across all moral systems – will have the intuitions that we should sometimes keep our agreements, sometimes help people, and that we should sometimes refrain from killing others.⁴⁷

Furthermore, some have argued that these moral claims are conceptual truths about morality. If you understand the concept of an unmarried man and a bachelor you will understand that unmarried men are bachelors. Terence Cuneo and Russ Shafer-Landau argue that, similarly, if you understand morality at all, then you will believe propositions such as the following to be true (so long as you understand them): it is pro tanto wrong to engage in the recreational slaughter of a fellow person; it is pro tanto wrong to break a promise on which another is relying simply for convenience’s sake; and there is some moral reason to offer aid to those in distress, if such aid is very easily given and comes at very little expense.⁴⁸ Even critics of Cuneo and Shafer-Landau’s proposal believe that conditional versions of these claims might be conceptual truths: for example, that it is a conceptual truth that if anything is pro tanto wrong, it is pro tanto wrong to engage in the recreational slaughter of a fellow person.⁴⁹ And if these conditional claims are conceptual truths, that would be enough for my purposes. For this would show that these intuitions are not ones we only contingently have: the intuition that bachelors are unmarried men is not an intuition we have only contingently about the concept of a bachelor; for if we didn’t have this intuition we wouldn’t have the (same) concept of a bachelor. Similarly, if these moral intuitions are intuitions of conceptual truths about moral concepts, then they are not intuitions that we only contingently have about the wrongness/rightness of these actions.⁵⁰

Question:

Shouldn’t ethical theories explain why certain intuitions are not widely shared or why certain moral views are controversial? It seems that it is incumbent upon ethical theories to explain why there are disagreements in intuitions about abortion, retributive punishment, and trolleys. But according to Shared by Almost All this isn’t the case. Doesn’t this make Shared by Almost All implausible?

Shared by Almost All just sets out the positive intuition-related desideratum on ethical theories. That is, which intuitions it is desirable for ethical theories to fit with and explain. It is consistent with this that a good ethical theory should also explain why certain moral issues are controversial or why there are 50/50 or 60/40 splits of intuitions about certain moral topics.

Question:

Isn’t Shared by Almost All very conservative? Doesn’t it entail that we shouldn’t accept revisionary ethical theories such as act-utilitarianism?

In fact, Shared by Almost All is relatively friendly to somewhat revisionary views in normative ethics such as act-utilitarianism. Act-utilitarianism might seem to fail to fit with and/or explain intuitions that are shared by almost all because it entails that we are morally required to give away a large proportion of our income to charity and almost everyone has the intuition that we are not morally required to do this. However, it is not obvious that almost everyone in fact has this positive intuition (rather than just not having the intuition that we are required to give so much to charity). And, even if almost everyone did have this intuition, act-utilitarians have a strategy that they can and do pursue that fits with what Shared by Almost All requires of them. Namely, act-utilitarians like Peter Singer argue that the view that we are not required to give away a large proportion of our income is in conflict with the view that we ought to save a drowning child in a pond when we can do so easily. If their argument is sound, then although act-utilitarianism may conflict with one intuition shared by almost all – that we aren’t required to give a large proportion of our income away – any view that does not conflict with this widely shared intuition will conflict with another intuition shared by almost all, namely that we ought to save the drowning child.⁵¹

The general strategy here is to argue that intuitions or views that most people share, when taken seriously and combined, yield a view that not everyone has or finds intuitive. This is a standard way of arguing for a revisionary ethical theory. And Shared by Almost All is perfectly consistent with arguing for ethical theories in this way.

It might be argued that almost all people believe that in no circumstances should a doctor cut up one healthy person so that he can distribute their organs to five people who will die unless they receive an organ transplant from this one healthy person: it would always be wrong for a doctor to do this. But act-utilitarianism entails that it is right for a doctor to kill the one to save the five in this way. So, if we accept Shared by Almost All, act-utilitarianism does not do well on the intuition-related desideratum on ethical theories.

It’s not clear that it would be bad for Shared by Almost All to entail that to the extent that it conflicts with our anti-transplant intuitions, act-utilitarianism does poorly on the intuition-related desideratum for ethical theories. According to Shared by Almost All, although act-utilitarianism does fit with many widely shared intuitions it doesn’t fit with our widely shared anti-transplant intuitions and that is a pro tanto vice of act-utilitarianism. This is a somewhat plausible view.

However, I think that an alternative view fits slightly better with our practice of ethical theorising. This alternative view holds that ethical theories can do well on the intuition-related desideratum so long as they fit with at least many of the ethical intuitions that almost all people who understand the relevant issues share. Consider

Shared By Almost All*. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains at least many of the ethical intuitions that almost all people who minimally understand the relevant issue(s) share. An ethical theory does particularly well on this desideratum if it either (i) fits with all the ethical intuitions that almost all people who minimally understand the relevant issues share or if (ii) for those that it does not fit with it shows that we have decisive reason to believe that these intuitions are off-track or unreliable.

Shared By Almost All* entails that act-utilitarianism can do very well on the intuition-related desideratum. Act-utilitarianism entails that we should keep our agreements, help others, and refrain from assaulting and killing people most of the time. This is because most of the time it’s better for us to do these things.⁵² So, it fits with many of the intuitions shared by almost all. If act-utilitarians can provide a very plausible error theory of the intuitions that are shared by almost all that act-utilitarianism conflicts with, then act-utilitarianism will do particularly well on the intuition-related desideratum. For it will fit with and explain all the intuitions that it needs to fit with and explain: the other intuitions will have been shown to be unreliable. This is exactly what several act-utilitarians try to do: they try to show that we have strong reason to accept an error theory of the intuitions that their view conflicts with. For instance, Joshua Greene argues on the basis of empirical studies that the deontological intuitions that many have about the footbridge trolley case do not track morally relevant considerations because the percentage of people who have these deontological intuitions significantly decreases when physical force and proximity are removed from the case; and whether the case involves physical force or proximity (whether we push the heavy man to his death or press a button from afar that drops him to his death) is not relevant to what we morally ought to do in these cases.⁵³

It doesn’t matter for my purposes whether Greene’s argument is successful or not. Regardless, if a theory has implications that conflict with some of the intuitions of almost all people, then proponents of that theory should pursue exactly the strategy that Greene pursues and develop an error theory according to which these intuitions are unreliable. This is what proponents of revisionary views like Greene and Singer already do. And Shared By Almost All* explains why they feel the pressing need to do this. Namely, because if they are successful in their attempts to provide error theories, then they will have shown that their view in fact does well on the intuition-related desideratum for an ethical theory. For they will have shown that the intuitions that their theory conflicts with are ones that – although held by almost all – are not ones that should be taken into account when we are considering whether an ethical theory fits with and explains the intuitions that it should fit with and explain.

Question:

Do Shared by Almost All and Shared by Almost All* take into account the intuitions that people held in the past?

I believe that I can be somewhat neutral on this issue. According to both principles, we can discount the intuitions of those in the past at least if we have strong reason to believe that their intuitions were unreliable. We do often have such a reason. For instance, we know that the intuitions of very many in the past were influenced by racist and sexist myths, ideologies, biases, and empirical claims based on these ideologies. Furthermore, note that the issue of whether we should take intuitions in the past or only in the present into account is an issue that all views about how controversial intuitions relate to ethical theories – especially all versions of Widely Shared—face, except Relativism. And in §2 I argued that Relativism faces other serious problems.

For the rest of this paper I will discuss Shared by Almost All rather than Shared by Almost All*. But this is only because: (a) the former principle is simpler, (b) the latter principle is really a specification of or addition to the former, and (c) the implications of these principles that I’m interested in in the rest of this paper concern how certain intuitions cannot be legitimately used to support explanatory ethical theories: the implications of these principles in this regard are the same. But I believe that Shared by Almost All* provides the more specific and accurate view.

5. Implications

Shared by Almost All entails that it is not a good-making feature of an explanatory normative ethical theory that it fits with and explains controversial ethical intuitions: that is, ethical intuitions that are not shared by almost all. This seems to mean that, seemingly contra Judith Jarvis Thomson and Frances Kamm, our intuitions about trolley cases cannot be utilized to motivate particular explanatory normative ethical theories (and argue against others).⁵⁴ Since, even the intuition that we ought not push the heavy man off of the bridge in the footbridge trolley case is not an intuition that is shared by almost all people.⁵⁵

W.D. Ross considers cases in which everything is equal between two worlds except that in the first world the virtuous are happy and the vicious are unhappy and in the second the opposite is true. According to Ross, we have the intuition that the first world is better than the second. And this intuition provides support for an objective-list account of value that holds that justice is one of the intrinsic goods.⁵⁶ But since it is not the case that almost everyone shares this intuition – many hedonists, desire-satisfaction theorists, as well as some objective-list theorists, do not share this intuition – we cannot rely on this intuition to motivate the view that justice is intrinsically good. I also asked the 196 participants that I recruited via Amazon’s Mechanical Turk platform (see §4) about a case like this. 46% of participants did not share Ross’s intuition about this kind of case.

Shared by Almost All also entails that retributive justice intuitions, such as the intuition that we should punish wrongdoers even if there are no good consequences to punishing them, cannot legitimately be used to support explanatory normative ethical theories. For such intuitions are not shared by almost all.⁵⁷ I asked the 196 participants that I recruited via Amazon’s Mechanical Turk platform about a case involving the punishment of a wrongdoer without any good consequences to such punishment and with close monitoring that did not constitute punishment as an alternative (all other things being equal between these alternative courses of action). 61% of respondents did not have retributive intuitions about this case, that is, did not have the intuition that a judge who could punish or monitor the wrongdoer should punish him.⁵⁸

Ross asks us to suppose that the fulfilment of a promise to A would produce 1,000 units of good for him, but that by doing some other act I could produce 1,001 units of good for B, to whom I have made no promise, the other consequences of the two acts being of equal value.⁵⁹ Ross reports the intuition that although breaking the promise would produce more good this could not justify us in breaking our promise and uses this intuition to argue for the view that it is pro tanto wrong to break a promise for its own sake even when there are no bad consequences to doing so. But it is not clear that everyone shares this intuition of Ross’s. Many consequentialists report not having this intuition.⁶⁰ I asked 77 people, whom I recruited from Amazon’s Mechanical Turk platform, about a case like this. 23% of respondents did not share Ross’s intuition about this kind of case. So, given Shared by Almost All, it is not clear that intuitions about cases like this can legitimately be used to argue for the view that it is pro tanto wrong in itself to break a promise or that there are non-derivative moral reasons to keep promises for its own sake regardless of the consequences of doing so.⁶¹

Does Shared by Almost All have any implications beyond normative ethical theorising? Does it have implications for theorising in political philosophy for instance? Controversial intuitions appear to play a large role in contemporary theorising about justice and equality. In arguing for sufficientarianism against egalitarianism, Roger Crisp relies on the intuition that if we must either give fine wine to 10 rich people or 10,000 super rich people, we should give the wine to the 10,000.⁶² In his response to Crisp, Larry Temkin reports the intuition that in this scenario we should either give no one the wine or at most give the 10 rich people the wine; we should not give the 10,000 the wine.⁶³ Temkin’s intuition – which is a kind of pro-levelling down intuition – is certainly not shared by almost all; and given Temkin’s response, perhaps enough people lack Crisp’s intuition such that it is not shared by almost all either. If theories of justice and equality are theoretically analogous to normative ethical theories, that is, if they are partially explanatory theories, then we should adopt an analogue of Shared by Almost All for theories of justice and equality and we should not rely on at least Temkin’s – and perhaps also Crisp’s – intuitions in our theorising about justice and equality: in which case these intuitions cannot legitimately be used to support (explanatory) theories of justice and equality.

Similarly, Elizabeth Anderson’s argument against luck egalitarianism relies on the intuition that justice and equality require tax-payers’ money or resources to be used to save the lives of reckless drivers without medical insurance who have caused an accident and will die by the side of the road without state sanctioned help.⁶⁴ I share this intuition of Anderson’s, but many political philosophers, such as libertarians, as well as some egalitarians, do not share this intuition.⁶⁵ And many members of the public – such as libertarians and the many who believe that we should treat the imprudent more harshly – do not share this intuition. So, it seems that the injustice of abandonment intuition is not shared by almost all.

So, if theories of justice and equality are explanatory theories in the way in which ethical theories such as rule consequentialism, contractualism, and Rossian pluralism are, then we should accept an analogue of Shared by Almost All for theories of justice and equality. And this analogue of Shared by Almost All will have significant implications for theorising about justice and equality. For it will entail that controversial intuitions, such as those discussed above, do not need to be explained by theories of justice and equality. And cannot be legitimately used to support theories, or objections to others, in (explanatory) theorising about justice and equality.

The question then is whether theories of justice and equality are partially explanatory theories. The early Rawls seemed to view his theory of justice as akin to explanatory ethical theories such as Ross’ and Sidgwick’s. Rawls compared his theory of justice to their ethical theories, and their theories were explanatory ethical theories.⁶⁶ And Rawls’ explicit remarks on the nature of moral theory and theories of justice seem to make it clear that he saw his theory of justice as an explanatory theory akin to an explanatory ethical theory. Rawls says now one may think of moral theory at first (and I stress the provisional nature of this view) as the attempt to describe our moral capacity; or, in the present case, one may regard a theory of justice as describing our sense of justice…’⁶⁷

Will Kymlicka claims that ‘there is a fundamental continuity’ between ethical theories and theories of justice.⁶⁸ Further, Kymlicka says that ‘the ultimate test of a theory of justice is that it cohere[s] with, and help[s] illuminate, our considered convictions of justice’.⁶⁹ And he understands how different theories of justice fit with our intuitions/considered convictions in a similar way to the way in which I characterized different ethical theories as fitting with and/or explaining the intuitions that are shared by almost all in §4. According to Kymlicka, ‘[d]ifferent theories appeal to our considered convictions in different ways. Utilitarians and libertarians, for example, answer to them in a more indirect way than liberals and feminists do … ’⁷⁰ Similarly, in §4 I argued that act-utilitarianism explains at least many of the moral intuitions that are shared by almost all but does so in a way that is less direct than the way in which its competitors fit with and/or explain those intuitions.

So, at least several influential political philosophers seem to have held that theories of justice and equality are akin to explanatory ethical theories. And if this is right, then we should accept an analogue of Shared by Almost All for theories of justice and equality, which would have significant implications for the intuitions that can be legitimately utilized in arguing for those theories.

Sometimes we do not know how many people share a particular intuition. A final implication of Shared by Almost All is that, for the purpose of normative ethical theorising, in these cases we should often gain a better understanding of how many people share a particular intuition. For we need to know which intuitions are shared by almost all in order to know which intuitions explanatory normative ethical theories need to fit with and explain.⁷¹ So, Shared by Almost All secures something of a nuanced role for experimental philosophy in explanatory normative ethical theorising.

In this paper I’ve argued that we should accept

Shared By Almost All. It is a desirable feature of an explanatory normative ethical theory that it fits with and/or explains the ethical intuitions that almost all people who minimally understand the relevant issue(s) share.

I argued that all alternatives to Shared By Almost All face severe problems but Shared by Almost All does not. And I argued that Shared by Almost All has interesting implications for ethical theorising, the role of experimental philosophy in ethical theorising, and for theorising about justice and equality.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1. See Nozick (1981, 363–397), Boonin (2008), and Bedau and Kelly (2015).

2. See, for instance, Dancy (2000, 168).

3. See, for instance, Singer (1975, 95–101)

4. See Hauser et al. (2007) and Kelman and Admati Kreps (2014).

5. On this issue see Wedgwood (2010) and Mogensen (2017).

6. See Shafer-Landau (2012, 15) and Hooker (2015, §7). Ross (1930, 21, 27) attempts to provide the most unified explanation of our moral duties. He reduces the number of his prima facie moral duties from seven to five because he comes to believe that our duties of justice and self-improvement can be subsumed under a more general duty to promote the good.

7. This is not to say that there is no connection between explanatory ethical theorising and arguments in applied ethics. But rather that there are several different projects that we can be engaged in when we are doing applied ethics and not all of these must have the explanatory aim that explanatory normative ethical theories have. For instance, ethicists engaged in one kind of project in applied ethics might take a certain controversial intuition as given and argue that given that intuition we should accept some other (controversial) conclusion. (e.g. it might be argued that if you have the intuition that animal cloning is morally permissible, then you ought to believe that human cloning is morally permissible too).

8. See Timmons (2013, 12–16).

9. To clarify: when I talk about theories fitting with and/or explaining the widely shared intuition that p I mean to be talking about them fitting with or explaining the content of the widely shared intuition that p.

10. This methodology is not too idiosyncratic. One persistent argument made in metaethics against relativist and reductive naturalist views of morality is that they entail that fundamental moral disagreement is impossible. But in our moral practice we engage in fundamental moral disagreements. So, such accounts might be giving a good account of something, but not of morality; see Olson (2014, 126–135). Relativists and reductive naturalists argue that in fact they can preserve much of our practice of seemingly genuine fundamental moral disagreement; see Finlay (2008). This debate seems to be one in which both sides take the aim to be to account for a feature of our moral practice. Similarly, in this paper I aim to give an account of the legitimate role intuitive support can play in our practice of explanatory normative ethical theorising that accounts for the way in which those engaged in this practice do and do not use (controversial) intuitions.

11. See Stratton-Lake (2016, esp. §1.1).

12. See, for instance, Haidt (2012) and Kim et al. (2012).

13. Kamm (2007, 8 n. 4)

14. See, for instance, Ross (1930/2001, 41), Hooker (2001, 14–16), and Timmons (2013, 14).

15. Norcross (2000, 141)

16. See Ross (1930/2002, 41), Kagan (2001, 45–46), DePaul (2006, 605), and Arvan (2016, 11–14).

17. This follows from the scientific commitment to common observation, replication, and repeatability; see Arvan (2016, 11–12) and, for instance, Gauch (2003, 11).

18. See Ross (1930, 41).

19. Arvan (2016, 14–15)

20. See, for instance, Hooker (2015, §7).

21. It has been put to me that consequentialists do not hold that their theories are justifiable to everyone because they fit with widely shared intuitions. However, first, even act-consequentialism is sometimes justified on the basis that it explains why many acts that we typically take to be wrong are wrong; see Shafer-Landau (2012, 125–126); rule-consequentialism is also justified in this way, see Hooker (2015, §7). Second, consequentialists do not justify their ethical theories on the basis of controversial consequentialist intuitions; they only try to make the counter-intuitive consequences of consequentialism seem more palatable in various ways; see, for instance, Singer (2005) and Greene (2013). This is distinct from claiming that consequentialism is positively justified partially because it fits with some consequentialists’ controversial consequentialist intuitions about cases.

22. Hooker (2001, 14–16) and Timmons (2013, 14).

23. See Hills (2009), McGrath (2009, 2011), and Archard (2011).

24. See, for instance, McGrath (2011).

25. Why must the methodology of explanatory normative ethical theorising be neutral on the controversial issue of whether there are moral experts? First, it is generally assumed that the right way of theorising in normative ethics involves adopting the method of wide reflective equilibrium. But wide reflective equilibrium involves trying to ensure that our best theorising in normative ethics is consistent with our best theorising in other domains such as social science, metaphysics, epistemology, and metaethics; see, for instance, Scanlon (2002, esp. 146) (2014, 21). Second, we should generally prefer theories that don’t have commitments that are controversial in other domains. Other things equal, we should prefer theories in epistemology, ethics, metaethics, and aesthetics that don’t depend on the existence of possible worlds for instance. Why? Suppose that all things are equal between theory 1 and theory 2 except that theory 2 entails a controversial metaphysical commitment. In this case theory 2 is less likely to be true. (This reason to prefer theories without controversial commitments is related to the reason to prefer more parsimonious theories). We can scale this up: if a practice of theorising relies on a controversial metaphysical commitment, then it is less likely to be a way of theorising that yields truths than alternative ways of theorising that do not rely on a controversial metaphysical commitment.

26. See Hursthouse (1999, ch. 1, esp. 28), Zagzebski (2004, 160), and Hursthouse and Pettigrove (2016, §2.2).

27. This problem might seem to motivate the view that it is a desideratum of a normative ethical theory that it fits with and/or explains the intuitions of the majority of the virtuous. But such a view would not help, since there is disagreement about who the virtuous are; see McGrath (2008).

28. See Jones (2017) and Moore (2015), and the YouGov and Gallup polls therein.

29. See Graham et al. (2009, 1031–1033) and Kim et al. (2012).

30. Jones (2017).

31. See Haidt (2012, ch. 1).

32. Given Haidt’s (2012, ch. 1) studies and my surveys discussed above.

33. It might seem that instead of Shared by the Majority we should accept

Scalar. A normative ethical theory is better or more strongly justified the more people’s ethical intuitions it fits with and/or explains. But Scalar encounters a similar problem to that which Shared by the Majority faces: it entails that an ethical theory is better and more strongly justified to the extent that it entails (1-3) and to the extent that it entails that whether an action involves (4-6) is relevant to the moral status of that action.

34. Foot (1978, xii)

35. Huemer (2005, 102)

36. Temkin (2012) argues that we should reject the content of the second, transitivity of value, intuition. But he doesn’t argue that we do not share the intuition that the transitivity of value holds.

37. See Cosker-Rowland (2017) which also responds to the well-known empirical arguments of De Brigard (2010) and Weijers (2014).

38. As noted in Cosker-Rowland (2017) all but one of the participants who did not have the anti-life-in-the-experience-machine intuition indicated that they had misunderstood the question or ticked the wrong option. With those participants removed, 99% of participants reported the anti-life-in-the-experience-machine. So, this data seems to show that 99% of those who understood the relevant issues had the anti-life-in-the-experience-machine intuition.

39. See, for instance, Rini (2015, esp. 434).

40. See, for instance, Kumar (2016).

41. See Arvan (2016, 12, 24–25).

42. See, for instance, Streumer (2017). Furthermore, as I explain below, error theorists have conditional moral intuitions such as that if anything is morally wrong, then recreational slaughter is. And that they have these conditional intuitions is enough for my purposes. See infra notes 49–50.

43. See Lackey (2008, 307–308) and Christensen (2010, 10).

44. See Johnson (2018).

45. See Oliphant (2018)

46. See Curry et al. (2019).

47. Empirical data may also give us good reason to believe that our intuitions about particular cases are not as fleeting as majority opposition to same-sex marriage and abortion is. The Schwartz value survey in psychology has been taken by over 40,000 people in over 53 countries. And the results seem to show that values beyond pleasure, such as social achievement and autonomy, are shared universally; see Alfano (2016, 162–165). In this case, we would expect to see stable pro-life-outside of-the-experience-machine intuitions shared by almost all across all cultures. It might be argued that anti-experience-machine intuitions are a product of a contingent distrust in technology. However, the results discussed above in which over 90% of participants reported anti-experience machine intuitions exclude cases in which participants reported an anti-experience-machine intuition but when asked why they had that intuition reported distrust in the technology of the experience machine; see Cosker-Rowland (2017).

48. Cuneo and Shafer-Landau (2014, 405)

49. See Evers and Streumer (2016).

50. Normative ethical theories need only take into account intuitions about what things would have particular moral properties if there were any moral properties. Similarly, if we are giving a theory of witches, we can utilize people’s intuitions about what witches would be like if there were any; not just people who believe that there are actual witches’ intuitions about what actual witches are like.

51. Or that the physical distance between us and someone whom we can save is not a morally relevant feature of a situation; see Singer (1972).

52. Shafer-Landau (2012, 15)

53. See Greene (2013, esp. ch. 9).

54. See Thomson (1985) and Kamm (2007).

55. Around 40% of Weijers et al. (2019) sample of 467, and Cosker-Rowland and Killoren’s (2020) sample of 114, judged that it was permissible or right to push the heavy man in the footbridge case; on the latter see Cosker-Rowland (2020, 211).

56. Ross (1930, 138)

57. See supra note 1.

58. The case that I asked respondents about was the following: It is 2200. Robert lives in New Zealand. He murdered someone and has been convicted of a murder. The murder is the first in New Zealand since 2100. During the conviction process Robert has been stably rehabilitated: he has practically become a new and different person following his arrest. With the new technologies that have been developed everyone knows that Robert will not commit a murder again. And whether he is punished for committing the murder won’t have any social consequences: if he is punished this will not deter future crimes, his not being punished will not encourage future crimes or murders. The judge in charge of sentencing Robert can either punish Robert by incarcerating him in a prison where he will be lonely and not very comfortable, restricting his diet, and not letting him do things that he wants to do. Or the judge can rule that Robert will be closely monitored in his house and his workplace at all times. Robert in fact wants to be closely monitored, and would not view this as a punishment, because he doesn’t want to ever commit a crime again. Should the judge punish Robert or monitor him?

59. ibid. 34–35

60. See, for instance, Singer (1975, 95–101).

61. The case that I asked respondents about was the following: Alice is very old and knows that she will die tomorrow. Her last decision is which charity to donate a sum of money to. Alice has promised Becky – who has no money of her own – that she will give $1000 to Becky’s favourite charity, charity 1. The maximum donation that charity 1 will take is $1000. There is another charity, charity 2, which is just as good and as effective a charity as charity 1. The minimum donation that charity 2 will take is $1001. Alice cannot give to both charities because she does not have enough money to do so. There are no other charities that Alice can give to. What should Alice do? If she gives $1000 to charity 1, she will keep her promise to Becky. If she gives $1001 to charity 2, she will break her promise to Becky. Neither Becky nor anyone else will ever find out which charity Alice gave the money to.

62. Crisp (2003, 755)

63. Temkin (2003, 771)

64. See Anderson (1999, 295–296).

65. See Nozick (1974, e.g. ix), Narveson and Sugden Sherwood (2003, esp. 419–420, 428, 431), and Huemer (2013, 148–159). Michael Otsuka is one egalitarian who seems to have indicated to me that he does not share this intuition; (p.c. 16 May 2010).

66. See Rawls, (1971/1999, ch.1).

67. ibid. 41

68. Kymlicka (2002, 5)

69. ibid. 6

70. ibid.

71. However, when the case is particularly complicated (e.g. some of Kamm’s cases) more interaction with respondents than the mere administration of a survey will be needed.