In this article I will outline a brief case for ethical naturalism, which is the view that morality is real and arises purely from aspects of the natural world. My argument will proceed in three parts. First, I will attempt to provide some conceptual clarity by outlining what we mean when we talk about morality, focusing on what sort of thing morality is and what it would entail if it existed. Second, I will sketch an example of a naturalistic moral theory, specifically a theory of reductive moral naturalism outlined by Peter Railton. Third, I will consider some objections to ethical naturalism: motivational internalism, the triviality objection, and the problem of normativity. I will argue that these objections do not substantively undercut the case for ethical naturalism.
What is morality?
Before we can answer the question ‘does morality exist’, we must first determine what we are even talking about when we say ‘morality’. At its most fundamental, morality is a code of conduct for human behaviour which specifies some actions (and inactions, attitudes, motivations, etc) as appropriate or proper, and others as inappropriate or improper. Behaviours congruent with the moral code are praised, while those incongruent with it are condemned. According to moral universalism, there is one privileged code of conduct which is applicable everywhere, at all times and in all societies (at least among humans; here for simplicity I will leave aside issues of animal and machine ethics). This privileged code of conduct, which we might call the ‘correct morality’, need not involve any very specific norms, but might consist of quite general standards and principles which could then be applied differently in different societies depending on circumstances. So for example the ‘correct moral code’ might specify that it is wrong to kill human beings without a very good reason and without proper due process, but exactly what constitutes a good reason and due process may well depend on the social precise circumstances.
In asking whether morality exists, therefore, we are asking whether there is a single correct code of conduct for human behaviour that is applicable to all human societies, even if those societies were unaware of it or chose to ignore it. Traditionally many have identified this privileged code of conduct with God’s laws or commandments. Ethical naturalism, however, is the view that moral facts are natural facts about the physical world, and not the product of some divine injunction or transcendent cosmic principle. Absent some sort of creator or other privileged supernatural being, what else could give rise to, or could account for the existence of, a privileged code of human conduct? Ethical naturalists are those who believe that it is possible for such a privileged code of conduct to exist purely in the natural world. How could this be the case?
To answer this question we must first observe that there is more to morality than simply being a privileged moral code of conduct. Though the details vary, there is effectively universal agreement that the privileged code of conduct that is morality necessarily promotes pro-social cooperation of people within a society, proscribes various behaviours that are detrimental to the self and to others, and promotes fairness and equity. Of course there is sharp disagreement about how to understand notions like ‘harm’ and ‘fairness’, but the point here is simply that the universal code of conduct referred to by morality, if such as thing exists, must relate in some central way to reducing harm and promoting fairness and equity. We could imagine other universal codes of conduct, but I argue they would not be a moral code of conduct, since our conception of morality necessarily and intrinsically includes these notions. This constraint is important, because it provides sufficient detail to begin constructing an account as to how morality could exist in a naturalistic universe.
A theory of ethical naturalism
Armed with a basic conception as to what we mean when we talk about morality, we are now ready to outline a theory as to how ethnical naturalism can account for the existence of moral facts. Here I will present only one of the many theories that have been developed, that propounded by philosopher Peter Railton. The details of his theory are not the major focus of this article, so I will offer only the briefest outline. The key idea is that moral facts derive from what would maximise the fulfilment of idealised preferences. An idealised preference is not what somebody actually wants, but what they would want themselves to want, if they had access to full information and were perfectly rational. This is important because people can want things that are bad for them (e.g. wanting to smoke). Railton’s account also holds that moral facts refer to what would maximally satisfy the sum of idealised preferences aggregated over all individuals, treating each person equally. Even though the idealised preferences of any given person may be purely selfish, the satisfaction of idealised preferences across all people necessarily incorporates the wellbeing of all persons, and thereby provides a basis for moral facts. Thus, according to Railton’s theory, an action is morally good to the degree to which it contributes to satisfying the sum of everyone’s idealised preferences, treating each person equally.
Railton’s account provides a naturalistic theory, because facts about idealised preferences are natural facts – they relate to things in the natural world (specifically idealised human desires about promoting human welfare), as opposed to divine commands, logical abstractions, or metaphysical principles. It is also very clearly a moral theory, since it provides (in outline form) a code of conduct for human behaviour relating in its very essence to human welfare and action in social contexts. Though I lack the space to make the argument here, I also believe that Railton’s theory of morality as maximising the fulfilment of idealised preferences provides an account of the universal code of social conduct that best fits our antecedent notion of what such a code we are looking for. That is, compared to any rival accounts, Railton’s best fits what we mean when we think and talk about morality. As such, just as in science we accept the theory that best accounts for and explains the available data, so too should we regard Railton’s theory as privileged over others, and at least as approximately describing the ‘correct morality’.
Parfit argues against idealised preference theories of morality on the basis that they provide no constraints on what people’s idealised preferences could be. He uses the example of an anorexic girl who, even after full reflection and access to all relevant information, could conceivably still decide that her idealised preference is for her to starve herself to death. According to Railton’s theory, this would mean that this would then be what is non-morally good for her. I personally do not consider this to be a strong objection to idealised preference theories of morality. This is because the idea that anyone who was fully rational and had access to all pertinent information regarding ways of living would still decide that their idealised preference would be to starve themselves to death is, in my view, totally absurd. This is precisely why conditions such as anorexia and depression are rightly regarded as mental disorders, involving false beliefs and a variety of cognitive distortions. If we begin with an assumption that even idealised fully rational, fully informed versions of ourselves would be subject to such defective views, then it does not surprise me that we arrive at absurd conclusions. I see this as stemming from the absurd initial assumption, and not from any particular flaw in Railton’s theory.
Motivational internalism objection
There are three major lines of objection often raised against theories of ethical naturalism. The first such objection comes from adherents of a view called motivational internalism, who argue that motivation is an essential component of morality. According to this view, a belief in some moral fact necessarily involves a motivation to act in accordance with that fact. For example, to believe that it is wrong to eat animals would necessarily entail a motivation not to eat animals. Perhaps that motivation would be overwhelmed by a stronger motivation, but nevertheless it must exist in some form. Anyone who did not possess such a motivation would, under this view, not truly believe that it was wrong to eat animals, even though they may protest to such a belief.
The reason why motivational internalism poses a problem is because it seems hard to fit into a naturalistic worldview. This is because under moral naturalism moral facts are simply natural facts, and it does not seem like natural facts are the sorts of things that necessarily lead to any particular motivation. We can believe all sorts of things about the solar system, the human body, societies, physics, etc, without any necessary motivational state being attached to or following from this. This idea that beliefs are never sufficient for producing motivations to act is known as the Humean theory of motivation, after its main populariser David Hume. According to Hume, for a motivation to exist we must have both some sort of antecedent desire, plus a belief that acting in a certain way would satisfy that desire. Mere belief itself is never sufficient to produce motivation. If we accept both motivational internalism and the Humean theory of motivation, it follows that when people come to believe in a moral fact, that belief always produces a relevant desire to act. Many philosophers regard this as hard to fit into a naturalistic worldview, as there just don’t seem to be any facts about the natural world that necessarily produce desires in this way.
My response to this issue is to simply reject motivational internalism is being too strong and too demanding a view. After all, why should we think that moral beliefs necessarily imply or generate a corresponding motivation? This does not appear to be the case empirically, since there appears to be strong evidence for the existence of psychopaths who know about morality but remain unmotivated to act in accordance with it. Furthermore, it also seems to be the case that the principle fails to apply to other fields of enquiry. We can, for instance, imagine recalcitrant persons who agree that an argument is sound and fail to identify any logical mistake with it, but nevertheless still have no motivation to accept the conclusion of the argument as true (indeed, many of us have likely participated in discussions where this behaviour has manifested!) Given these considerations, I do not see any strong reason to accept motivational internalism, and as such the failure of natural moral facts to necessarily supply any motivation to act in accordance with them does nothing to undermine the case for ethical naturalism.
The triviality objection
Parfit raises a second objection to ethical naturalism, which he calls the ‘triviality objection’. According to this argument, it is impossible for ethical naturalists to simultaneously argue that ethical facts are natural facts, and also maintain that this is a substantive, informative claim, more than just a mere tautology. The example that he uses considers two properties: the natural property ‘maximises happiness’, and the moral property ‘is what we morally ought to do’. Ethical naturalists argue that a property something like ‘maximises happiness’ is the same as the property ‘is what we morally ought to do’. Yet according to Parfit, these two properties cannot simply be identical, as otherwise there would not be two properties but one, and we would essentially just be saying ‘property A is property A’, which is an uninformative tautology. To be a substantive claim, ethical naturalists must instead be saying ‘property A is property B’, but Parfit doesn’t think this makes any sense, since there must be something to distinguish the two properties for them to be different. He gives the example of water and H20, arguing that although water and H20 are the same substance, the property ‘is comprised of two hydrogen atoms and one oxygen atom’ is not the same as the property ‘is a clear substance that falls from the sky and we need to drink to survive’. These two properties might be satisfied by the same stuff, but the properties themselves are not the same. Likewise, Parfit argues that states of affairs can possess both natural and moral properties, but these properties will always and necessarily be distinct and different properties, not the same as the ethical naturalist claims. Parfit thinks this is an argument for irreducible moral properties, which cannot be reduced to or equated with natural properties.
The flaw in Parfit’s argument, in my view, is that he does not articulate what it means for two apparently distinct properties to be the same. We can address the apparent paradox by appealing to the distinction between sense (the internal psychological meaning of a phrase) and reference (the thing in the real world picked out by a phrase). The classic example of this is that of the ‘morning star’ (a star that is visible in the east just before sunrise), and the ‘evening star’ (a star that is visible in the west just after sunset). Although the phrases ‘the morning star’ and ‘the evening star’ do not have the same meaning, they in practise refer to the same thing, namely the planet Venus. We can apply this example to response to Parfit’s concern. The natural property ‘maximises happiness’ does not mean the same thing as the non-natural property ‘is morally good’. However, it turns out that both properties are ‘the same’, by which I mean that:
- All pertaining states of affairs with property A also have property B
- There are facts that account for this coincidence of properties, so it is not simply an accident
I argue that this is what reductive naturalists mean when they say that the property ‘is morally good’ is the same thing as the property ‘is morally good’. Understood in this way, there is no paradox about how two properties can be the same and yet different. They are different in that they mean different things (have different senses), but are the same in that they have the same referent (the properties are fulfilled by the same states of affairs and only those states of affairs). The reductive naturalist’s claim that the two properties are the same is therefore substantial, and not trivial. As such, I believe that Parfit’s triviality objection fails.
The problem of normativity
The third objection to ethical naturalism, and probably the most important, is the problem of normativity; relating to the ‘binding force’ of morality. Philosophers typically understand this to mean that all persons have reasons to act morally, even if they may be unaware of those reasons or choose to reject them. This idea is called moral rationalism, the view that if it is wrong for somebody to do a particular act, then there is always a reason for them not to do that action. We would, for instance, typically accept the claim that smokers have a reason to quit smoking (namely the health benefits), even if they are unaware of those benefits or simply choose to ignore them. Likewise, it is argued that all persons have reasons not to kill, steal, etc, even if they fail to act in accordance with those reasons. As with motivational internalism, many thinkers have regarded normativity as a problem for ethical naturalism, on the grounds that natural facts simply are not the sorts of things that necessarily give rise to any particular reason to act. This is related to J. L. Mackie’s famous ‘argument from queerness’, in which he argued against the existence of moral facts on the basis that they would need to posses some queer property of being intrinsically motivating, or have an ‘ought-to-be-doneness’ about them. This sort of property seems hard to fit into a naturalistic worldview, because natural facts simply describe the way things are in the world. The way things are, however, doesn’t impose any obligations on us, or provide any reasons to act one way or the other. According to this argument the ethnical naturalist is, therefore, unable to account for the normativity of morality.
The challenge, then, comes down to what can be said in answer to the amoralist? This is a person who recognises the existence of facts about idealised human preferences (in line with Railton’s account), and perhaps even agrees that this provides the best account of a code of conduct pertinent to moral issues, but nevertheless demands to know why they have a reason to act in accordance with this code. In my view, the best response to this challenge is to argue that it is a basic, foundational principle that in any domain, rational agents have reasons to act in accordance with the privileged code of conduct (if any) pertinent to that domain. To understand how this answer works, suppose somebody were presented with a mathematical proof which they understood and followed in every stage, but then simply refused to agree that they had any reason to accept the conclusion. We could imagine them retorting that they have no internal desire or motivation to form accurate beliefs about this mathematical question, and therefore reject the idea that they have a reason to accept the conclusion of this proof. I would content, however, that regardless of their particular desires, the person in question still has a reason to accept the conclusion of the proof, because that is consistent with the privileged norms of rational inference governing the pertinent domain (in this case mathematics). To take another example, we would typically say that anyone engaged in a game of chess would have a reason to make a move that would help them to win the game – even if they did not feel any desire to make the move, and even if they didn’t care about winning the game at all. The fact that the person is engaged in the domain of chess means that there is a privileged code of conduct pertinent to that domain which provides them with some reasons for action, irrespective of their desires or motivations. The recalcitrant person in such cases is unlikely to be persuaded by this argument, but it seems to me that it correctly describes the reason they have for accepting the proof or making the chess move.
I argue much the same thing applies for morality. The domain of conduct pertinent to morality is that of living and interacting with other people. Unlike a game of chess or a mathematical proof, there is no real way to avoid engaging in the moral domain. Even if we become a hermit and cut off contact from others, there would still be other people in the world who would come under the remit of the domain of morality (since it applies universally across all people). People may fail to be motivated to act in accordance with what will best promote the wellbeing of themselves and others, but nevertheless they have a reason to act in accordance with this code of conduct, since that is the privileged code of conduct for the domain of living and interacting with other people. This might not be a very satisfying answer to the problem of normativity, but this will be an issue in any domain, since regardless of what reasons have been given for doing something or believing something, I can always still press the question ‘but why?’ Our chain of justifications has to stop somewhere, and I believe it is reasonable to affirm as fundamental the principle that everyone has a domain-specific reason to act in accordance with the privileged code of conduct pertinent to a given domain.
Thus, since everyone has a body that can be healthy or diseased, everyone has a health-related reason to stop smoking even if they don’t care about their health. We would not accept that a person who doesn’t care about dying of lung cancer actually has no reason to quit smoking; rather we would instead say that they are not motivated to act in accordance with this reason. Since (almost) everyone has some sort of money or property, (almost) everyone has a finance-related reason to save and invest their money wisely even if they don’t care about money. Since everyone lives in some sort of society with conventions of how to behave, everyone has an etiquette-related reason to obey their society’s rules of etiquette even if they don’t care about being polite. Likewise, since everyone lives on a planet where their actions potentially affect other people, everyone has a morality-related reason to act morally, even though they may sometimes fail to be motivated by these reasons. This is what provides the basis for the normativity of morality.
Glossary of some key terms
- Moral realism: there is a privileged, universal code of conduct governing human action, which gives rise to moral facts
- Reasons internalism: having a reason to do something implies having a motivation to do that thing
- Humean theory of motivation: beliefs are insufficient for motivation, need antecedent desire too
- Moral rationalism: if something is morally wrong then there must be a reason not to do it
- Ethical naturalism: moral facts are natural facts
- Reasons internalism + Humean theory: having a reason to do something implies a belief and an antecedent desire
- Reasons internalism + Humean theory + moral rationalism: to believe that something is wrong must necessarily produce or elicit a desire not to do that thing
- Reasons internalism + Humean theory + moral rationalism + ethical naturalism: there are natural facts, belief in which necessarily produces or elicits a desire to act in accordance with the universal code of conduct which governs human action = Mackie’s ‘categorically prescriptive facts’: facts that provide reasons for action independent of our desires