Moral affordances and the demands of fittingness

ABSTRACT

Some situations appear to make moral demands on us – they call for a certain response. How can we account for such paradigmatic moral experiences? And what normative properties or relations are involved? This paper argues that we can account for such moral experiences in terms of moral affordances, where moral affordances are opportunities for fitting action. The paper demonstrates that the concept of affordances helps to generate new insight in moral inquiry, especially in relation to the moral significance of fittingness.

KEYWORDS:

• Affordances
• fittingness
• moral affordances
• moral action
• fitting action
• meta-ethics

1. Introduction

Some situations appear to make moral demands on us – they call for a certain response. How can we account for such paradigmatic moral experiences? And what normative properties or relations are involved? In this paper, I argue that we can account for such moral experiences in terms of moral affordances, where moral affordances are opportunities for fitting action.

Affordances, in general, are opportunities for action (Gibson, 2014). Affordances have been extensively explored by both psychologists and philosophers in the study of action and action perception (e.g., Chemero, 2003; McClelland, 2019; Rietveld & Kiverstein, 2014; Siegel, 2014; Stoffregen, 2003; Turvey, 1992). But the moral dimension of affordances remains underexplored.¹ Yet, as already highlighted by Gibson, affordances can be beneficial or harmful (2014: 119, 129). Food in the fridge might afford nourishment, but if it has gone off, it might also afford poisoning. The benefits and harms of affordances are often morally neutral. A chair might afford – even solicit! – resting, but whether you take up this opportunity is (normally) morally neutral. Some affordances, however, are not morally neutral. Some offer opportunities for immoral action. Others pick out moral actions. The latter I call moral affordances. To illustrate, a sincere apology that you receive might afford forgiving, and you might experience the situation as calling for your forgiveness. By shedding light on the relation between given circumstances and the opportunities for moral action that they present to us, moral affordances allow us to make sense of such experiences.

What distinguishes moral affordances from other opportunities for action? I argue that moral affordances are opportunities for fitting action. A moral affordance, in other words, is the appropriate response to a given situation, and which can be experienced as being called for by the situation. Fittingness is receiving increasing attention in moral philosophy, especially in meta-ethics, but what fittingness is, and how it relates to moral action, remains unclear. I will show that using fittingness to develop the idea that (some) moral experiences are of moral affordances presented by a particular situation clarifies the role of fittingness in moral action.

In sum, my overall aim in this paper is to show that affordances lend themselves to play a bigger role in ethical inquiry than they have so far. Moral affordances help us make sense of some paradigmatic moral experiences as experiences of fittingness, thus shedding fresh light both on moral phenomenology and on fittingness as a basic normative property.

The paper makes three main claims. The first, descriptive, claim is that we can account for some paradigmatic moral experiences in terms of moral affordances and that moral affordances are a specific subset of affordances. The second, normative, claim is that the demands that are being experienced in an experience of moral affordances are the demands of fittingness. The final, comparative, claim is that moral affordances offer a better account of these experiences than alternatives – in particular, accounts involving reasons or obligations.

The paper is organized as follows. Section 2 offers a brief characterization of the moral phenomenology of direct moral demands. Section 3 introduces affordances, in general, and explains their suitability for investigating moral experiences. Section 4 explains my proposal that moral affordances, understood as a subset of affordances, are opportunities for fitting action. Section 5 considers alternative accounts of the phenomenology of direct moral demands. In section 6 I further develop my account by defending it against the alternatives. Section 7 concludes.

2. The phenomenology of direct moral demands

Moral phenomenology provides us with a starting-point for meta-ethical inquiry. What are moral experiences like? And, given what they are like, how can we account for such experiences?² Maurice Mandelbaum (1955), in his rich discussion of moral phenomenology, has identified two key features of paradigmatic moral experiences: that some situations appear to make direct demands on us and that the source of these demands appears to be outside of us, or independent of us.³

Suppose you see a child falling into the pond right in front of you, and you’re the only bystander who can help. In this situation, you’re likely to experience a demand to rescue the child and you’re likely to judge, based on this demand, that you should try and rescue the child. Moreover, this demand doesn’t just appear to be an inner demand. It isn’t merely the experience of a desire you have to help the child, for example, although such a desire might be present, too. Instead, it is felt as having an authority that is independent of your attitudes, as being what the situation demands or calls for.

Similarly, suppose you’ve just received a heartfelt apology from a longstanding and dear friend for something they did that was quite out of character. In this situation, you’re likely to experience a demand to forgive, even if you’re still hurt and perhaps still feel some resentment. You might not find it easy to forgive, but you’re also likely to feel that the demand to forgive has an independent authority over you.

I take it that Mandelbaum was right to highlight that experiences of this kind – call them experiences of direct moral demands – are paradigmatic moral experiences. But how can we account for such experiences? This is not an easy question to answer. And before I address this question, some clarifications will be useful.

First, we should note that the phenomenology of direct moral demands is only one aspect of a comprehensive moral phenomenology. Other important moral experiences include deliberation about the moral rightness and wrongness of different responses to a given situation, and, more generally, deliberation about moral rightness and wrongness and other moral properties. The phenomenology of judging the moral actions of others is also likely to be distinct from the phenomenology of responding to direct moral demands.⁴ I’ll bracket those less immediate moral judgments here, to focus only on direct moral demands. The experience of direct moral demands is an important, but underexplored, part of moral phenomenology and difficult to account for.

Second, I take the experience of direct moral demands to still leave us with the need to make a choice. The experience is not merely of being in the grip of an instinctive or habitual reaction to some environmental stimulus. In the pond example, you could decide not to jump. And you might have other motivations for jumping – for example, to impress a friend. But it’s also possible – and likely – that you jump because you experience that rescuing the child is what the situation demands from you. And we need to understand better what the basis is for such a decision.

Third, a satisfactory account of the phenomenology of direct moral demands must deliver on two fronts. It should give us an account of both the content of those experiences and the modality of those experiences. What is being experienced when we experience direct moral demands? And how are direct moral demands experienced? I argue that moral affordances, understood as opportunities for fitting action, have the potential to deliver on both fronts. My main emphasis in this paper, however, will be on the first question: how can we account for the moral demands that are being experienced?

A final clarification concerns the question what such experiences imply about the metaphysical status of the normative properties or relations that are being experienced. I’m not assuming that all such experiences are veridical, but I’m taking it as my starting-point that at least some of those experiences are veridical. It’s of course possible that these are just experiences and that the correct explanation of the experience of direct moral demands is entirely psychological and does not involve any moral properties or relations. If this were the case, it would imply an error theory about the content of those moral experiences. An error theorist would hold that while there are moral judgments that appear to be a response to a demand with independent authority, such moral judgments are systematically false because such demands to not exist (e.g., Joyce & Kirchin, 2010; Mackie, 1977). Doubts about the possibility of accounting for such demands within an otherwise plausible meta-ethical view have drawn some philosophers to error theories. As my aim in this paper is to argue that we can account for the experience of direct moral demands, I will bracket error theory. Instead, I focus on positive proposals for how to account for this experience, and I’m assuming that at least some of those experiences are veridical. They are experiences of some normative properties or relations, and the question is, which ones.

Even if we assume that some of those experiences are veridical, we shouldn’t jump to conclusions about the nature of moral reality, however. In particular, we shouldn’t just assume that the moral phenomenology I described commits us to some form of non-naturalist moral realism. Other views of moral reality, including some form of naturalism, might also be able to account for the experience of an independently authoritative moral demand. Terry Horgan and Mark Timmons have made this point in a series of recent papers (2007, 2008, 2017, 2018). They argue that the presupposition that accepting the moral phenomenology I described entails some ontological commitment is widespread and shared among both error theorists and non-naturalist moral realists. Against both of those views, they defend a non-entailment claim, which holds that the experience of a moral demand with independent authority, while significant, need not have “ontological purport” (Horgan & Timmons, 2018, p. 325): the experience of a moral demand with independent authority doesn’t entail that the experience involves non-natural normative properties or relations.

I accept their non-entailment claim. As they also acknowledge, however, it doesn’t follow from non-entailment that we shouldn’t try to account for this moral phenomenology. The non-entailment claim only holds that we can’t corroborate the reality of these experienced demands through introspection on those experiences.⁵ It challenges us to show that, given our moral phenomenology and given non-entailment, there is an independently defensible meta-ethical view that can account for the experience of moral demands with independent authority. I believe this challenge can be met. However, instead of starting with a comparison of different meta-ethical views – realism versus anti-realism, and non-naturalism versus naturalism – I believe it’s helpful to approach this challenge focusing, first, on developing a positive account of the experience of direct moral demands and address metaphysical and epistemological questions on that basis.

So, to restate the main question this paper focuses on: how can we account for the phenomenology of direct moral demands? What normative properties or relations are involved? And how are they experienced? My aim is to show that we can account for this phenomenology in terms of moral affordances, understood as opportunities for fitting action. At the descriptive level, this proposal involves affordances. At the normative level, the proposal involves the normative relation of fittingness. In what follows, I’ll first explain core components of the proposal, before arguing that moral affordances give us a better account of the phenomenology of direct moral demands than existing alternative accounts.

3. Affordances

Moral affordances are a subset of affordances, in general. James Gibson (2014) coined the term “affordance” to capture the features of the environment that present opportunities for action – for human and non-human animals:

The affordances of the environment are what it offers the animal, what it provides or furnishes, either for good or ill. The verb to afford is found in the dictionary, but the noun affordance is not. I have made it up. I mean by it something that refers to both the environment and the animal in a way that no existing term does.

(2014: 119, emphasis in original)

For example, a chair might afford sitting on, a wall might be climbable, and a lake might afford swimming in. Affordances, according to Gibson, are always relative to some animals – human or non-human. A chair affords sitting on for (most) humans, but not for babies or elephants. A wall that is climbable for an expert climber might not be for someone who has never done any rock climbing.

While affordances are thus relative to some animals, they are not subjective. They are not merely projections of our desires, say. Affordances are offered by features of the environment – by particular objects, such as a rock, or by a situation as a whole. But they are not entirely objective either. As Gibson writes, “an affordance is neither an objective property nor a subjective property; or it is both if you like. An affordance cuts across the dichotomy of subjective-objective…” (Gibson, 2014, p. 121).

Affordances are taken to be perceivable (more on this in a moment), but to exist independently of being perceived (Gibson, 2014, p. 119). A rock, say, is not merely perceived as offering an opportunity to sit on. You can actually sit on it, and it affords sitting on even if you fail to perceive this affordance. But because affordances are opportunities for action, they are not just objective either, at least not in the sense of being categorical properties of certain objects or situations.

Affordances, as they are commonly understood, are relations between human or non-human animals and their environment, where the relation determines possible actions (Chemero, 2003; Prosser, 2011; Stoffregen, 2003).⁶ On this relational view, which I shall adopt here, affordances are neither properties of the environment nor properties of animals. Instead, an affordance is a three-place relation between an animal, their environment, and possible actions that are available to them in that environment. Because action possibilities are the result of an interplay between properties of animals and properties of the environment, they are neither objective nor subjective.⁷

With a bit more precision, we can say that the general form of affordances is this: environment e affords doing x for an animal a. So an affordance A can be described by a relation A(x, a, e). In the example of a wall that is climbable, the wall is the relevant feature of the environment (e), and it affords climbing (x) for a, an average climber.⁸

Affordances, as they are commonly understood, are a promising conceptual framework for the inquiry into the phenomenology of direct moral demands. They have some features that are structurally similar to key features of the experience of moral demands. Specifically, I want to highlight three such features. The discussion of these features should also further clarify the concept of affordances and its usefulness for moral inquiry. In what follows, I shall only focus on affordances for human agents, and bracket affordances for non-human animals.

A first feature is that affordances, as we saw, are action possibilities created by a given environment for a given agent. In an experience of direct moral demands, what is being experienced is, similarly, a response open to the agent in a given situation – say forgiving, or rescuing the child. The experience is neither just of features of the agent – their desires, say – nor just of objective features of the environment. As we’ll see below, trying to account for the experience of direct moral demands in terms of either subjective features of agents or objective features of the environment creates difficulties for existing proposals. In light of these difficulties, the shift to action possibilities created by a relation between an agent and an environment offers a promising alternative, worth exploring further.

A second feature of affordances that is relevant for the topic under investigation is that affordances aren’t limited to action possibilities offered by the physical environment, such as particular objects in that environment, for example. Some affordances arise from the social environment – other people – and from (social) events.⁹ For example, a gap in traffic affords crossing the road. My earlier example of forgiving is also an opportunity for action presented by a social event – an apology that you’ve received. This is not to say that the physical environment is irrelevant for direct moral demands. Indeed, as my pond example illustrates, the physical environment is importantly involved in a range of moral experiences. But moral experiences normally also involve other people, or other beings, and affordances lend themselves to investigate the action possibilities that arise from our interactions with others.

A third feature of affordances is that they can be directly perceived. That is to say, affordance theory holds that action possibilities can be experienced without being represented (Davis, 2020; Davis & Chouinard, 2016; Gibson, 2014, p. 119; Michaels & Carello, 1981). This feature of affordances has proven particularly important for the uptake of affordance theory in engineering and design. A user-friendly phone, for example, is designed in such a way that its key control functions – its affordances – can be perceived without requiring inference or memory or other representational processes. Moreover, while some affordances are perceived as inviting action, others appear involve some form of solicitation. They may be perceived as demanding action (Gibson, 2014, p. 130) or to request an action (Davis, 2020; Davis & Chouinard, 2016). In the philosophical literature, Susanna Siegel (2014) draws a helpful distinction between mere affordances and soliciting affordances. The former present us with opportunities for action. The latter are experienced as “soliciting, inviting, or otherwise prompting” a certain response (Siegel, 2014, p. 54). As Siegel argues, solicitation comes in degrees. What she calls “experienced mandates” are affordances that come with a strong sense of perceived solicitation, such that a particular response is experiences as being mandated – or called for – by the environment.¹⁰

Siegel doesn’t mention moral experiences in her discussion of mandated experiences. But it’s easy to see how this feature of affordances is worth exploring further in the context of the experience of direct moral demands. As I mentioned earlier, in this paper I’ll focus primarily on the content of the experience of direct moral demands rather than on the question of how such demands are experienced. But I wanted to highlight that both features of affordances – that they can be directly perceived, not requiring representation of an action as having certain properties, and that they can be perceived as soliciting action – are promising for an account of the experience of direct moral demands. I’ll briefly return to this point in the final section of this paper. In the next two sections, I concentrate on the content of the experience of direct moral demands.

4. Moral affordances as opportunities for fitting action

As we saw in the previous section, affordances are relational properties of an environment – an object or a situation – that present an agent with a possible action. In some cases, actions are perceived as being mandated, or called for, by the object or situation. I now want to argue that there is a subset of affordances that picks out moral actions, and that helps us make sense of what we experience when we experience direct moral demands. The affordances in this subset I call moral affordances. My focus from now on will be primarily on the affordances of a situation as a whole, rather than on affordances of objects, as moral affordances typically are a subset of the former.

I mentioned in the introduction that affordances can be beneficial or harmful (Gibson, 2014, p. 129). More generally, affordances are inherently value-laden, or normative (Jayawickreme & Chemero, 2008), because they describe the meaningful environment. As Gibson (2014: 131f) puts it:

The perceiving of an affordance is not a process of perceiving a value-free physical object to which meaning is somehow added … ; it is a process of perceiving a value-rich ecological object.

This sense in which affordances are value-laden or normative is tied to their meaningfulness to us. It doesn’t capture their moral value or moral normativity. For example, that a wall is climbable – that it affords climbing – is loaded with a sense of how things should go if all goes well. But that “should” is not the moral sense – it doesn’t imply that it would be morally right to climb the wall.

Jayawickreme and Chemero (2008) have used this point to argue that moral inquiry can’t be directly based on affordances. Instead, it should be based on a “moral analogue” of affordances (2008: 122), not on affordances as such. They are right to highlight the discrepancy between the inherent normativity of affordances and moral normativity. But instead of resorting to analogues, I want to argue here that moral inquiry can be directly based on affordances, although only a subset of them. This subset I call moral affordances.

Let me start by noting that many affordances are morally neutral. When out hiking, for example, that the river’s narrowness affords jumping across is morally neutral – everything else equal, it makes no moral difference whether you use the bridge to cross the river or jump. Other affordances are not morally neutral. The openness of someone’s purse might afford taking the money that is inside. In some Robin Hood type situations, this is the right thing to do. In other situations, it would be wrong.

In some circumstances, it’s possible that all action possibilities are morally neutral. For example, if we focus very narrowly on the action possibilities afforded by your new pen, it might turn out that all of them are morally neutral. You can use it to take some notes or to doodle, and, let’s assume, it makes no moral difference what you do. In other circumstances, however, not all action possibilities will be morally neutral, and there is at least one right response. The pond case illustrates this possibility. Beyond jumping in to rescue the child, you might also have the option to just walk away, or to go and find someone else who can help. But, let’s assume, the right response is to rescue the child. That response is the moral affordance of these circumstances.¹¹

Affordances, in general, pick out opportunities for action for an agent in a given environment without specifying the moral status of those action possibilities. But we can ask, among the actions that are possible in a given environment, is there a right response to the situation? That’s the question that moral affordances offer an answer to. The basic idea is that the right response is realizing the moral affordances of a given situation, rather than any of the other action possibilities of this situation.

Moral affordances, understood in this way, are a subset of affordances. But what distinguishes moral affordances from other affordances? What is the right response to a given situation? My proposal is that moral affordances are opportunities for fitting action. I’ll say more about how I interpret fittingness in a moment. But, to focus on the form of moral affordances for the time being, a moral affordance is a fitting action x_f available to agent a in circumstances c (their environment). Moral affordances can thus be described as three-place relations MA(x_f, a, c). In a Robin Hood situation (c), the moral affordance is for Robin Hood (a) to take the money and give it to the poor (x_f). For moral affordances, I prefer to use the term circumstances, rather than environment, to highlight that moral affordances are typically not offered by physical objects, but by features of a situation as a whole, including social features. This also brings the terminology closer to the one used in moral philosophy.

As fittingness is the normative element that distinguishes moral affordances from other affordances, we now need to clarify this component. Fittingness, understood as a basic normative property or relation, is receiving a lot of attention today (e.g., Howard & Rowland, 2022; McHugh & Way, 2022), but there are many different interpretations of fittingness, and a lot of open questions about what fittingness is and where it applies.¹²

We can roughly describe fittingness, as I understand it, as the relation in which a response stands to the circumstances it is a response to, when the response is merited by the circumstances.¹³ For example, praise might be fitting when someone has done something praiseworthy, but it is not fitting if you’re trying to gain an advantage. After a misstep, an apology might be fitting. Apologizing for something you haven’t done, just because it appears to be expected, is normally not fitting. I understand relevant responses as actions broadly construed here – as including both actions, narrowly construed, and attitudes.

Given this initial rough characterization of fittingness, we can say that the moral affordance of a given situation is the opportunity to respond appropriately to this situation. If a moral affordance is being experienced, the experience is of a response that is appropriate in the situation – of the response that is being called for by the circumstances. This proposal, as I will argue in a moment, has the potential to offer us a very promising account of the phenomenology of direct moral demands, which is preferable to existing alternatives.

Before I can further develop the positive proposal, however, I need to address an objection that my interpretation of fittingness invites, even from philosophers otherwise sympathetic to fittingness as an important normative property or relation. Addressing this objection will also allow me to further clarify my understanding of fittingness. Most contemporary work has tended to understand fittingness in a way that restricts its application to attitudes (Maguire, 2018; McHugh & Way, 2022; Rowland, 2022; Schroeder, 2010).¹⁴ Fittingness, however, can only help us make sense of the phenomenology of direct moral demands if it also applies to actions, not just to attitudes. As we saw, some paradigmatic moral experiences involve attitudes – e.g., forgiving. But many such experiences involve actions – such as the action of rescuing the drowning child.

Can actions be fitting? Or is my proposal doomed because only attitudes can be fitting? Interestingly, and in some contrast to much contemporary work, early philosophical work on fittingness understood it in a way that applies to both actions and attitudes (Broad, 2014; Ewing, 1939, 2012). Part of the issue here is how to interpret the standard of fittingness: what makes a response fitting? On one popular view, the standard of fittingness is internal to types of responses. On this internal standard view (see McHugh and Way 2016, for example), a fitting response is one that is merited – or correct – by the standards for this kind of response. Admiring a saucer of mud, to use a well-known example, is not fitting, because the attitude, in this case, wouldn’t meet the internal standard for admiration. If this were the right view of the standard of fittingness, it is difficult to see how fittingness would apply to actions, as actions don’t have substantive internal standards of correctness.

But I think we should reject the internal standard view, even for attitudes. In some cases, it’s meaningful to say that while an attitude satisfies an internal standard, it’s still not fitting. For example, suppose someone very close to you has died, which, on an internal standard view, suggests that being sad is fitting. But the overall situation might be such that sadness, while understandable, is not fitting. For example, suppose you’re responsible for small children, who do not understand the situation, and who would feel abandoned. In this situation, setting aside your feelings of sadness might be fitting. Similarly, if a good friend has achieved something that you’ve so far failed to achieve, you’re likely to feel envy. But envy might not be fitting in those circumstances. It could be that what is fitting, given your friend’s efforts, is wholehearted happiness at their success.¹⁵

The point generalizes. As Christopher Howard and Stephanie Leary (2022) have also argued, the internal standard view of fittingness struggles to accommodate normative authority in the right way. What is fitting is the response that a situation demands. This line of reasoning suggests an external standard view of fittingness. According to an external standard view, the fittingness of a response depends on the relation between a response and the situation it is a response to, not (just) on the type of response that it is.

If we adopt an external standard view of fittingness, then there is no reason to think that only attitudes can be fitting. Actions, narrowly construed, can be fitting, too. In what follows, when I talk of fitting action, I mean responses broadly construed, to include both attitudes and actions, narrowly construed.

To summarize, moral affordances, on the view that I propose, are normatively loaded relations between an agent, the circumstances they find themselves in, and responses open to that agent in those circumstances. They are a subset of affordances (descriptive claim), and they pick out fitting actions (normative claim).

5. Reasons and obligations

Having outlined my positive proposal, in this section, I consider alternative accounts of the phenomenology of direct moral demands. On the fittingness-involving account that I’ve developed so far, the experience of a direct moral demand is the experience of a moral affordance, understood as an opportunity for fitting action. The account is based on fittingness as a basic normative relation. There are two main alternatives to this fittingness-involving account. On the first proposal, what is being experienced is a moral obligation. On the second, the experience is of a moral reason to act in a certain way. In this section, I discuss these two alternatives. In the final section of the paper, I argue that the fittingness-involving account is superior to these two alternatives. This comparison will then also allow me to say more about my understanding of moral affordances and the fittingness of actions.¹⁶

A popular view in moral philosophy today is the reasons-first view, which attempts to account for all of normativity (and the evaluative, too) on the basis of normative reasons. But independently of whether reasons are the only fundamental normative properties or not, can we make sense of the moral phenomenology under consideration on the basis of reasons? The pond example that I used earlier might suggest such an interpretation. On a reasons-involving account, that the child has fallen into the pond is a normative reason for rescuing the child. Can that reason be the basis for an experience of a demand with independent authority? A positive answer to this question has been influentially defended by Jean Hampton in her book on The Authority of Reason (1998). Hampton’s account has the merit of engaging closely with the phenomenology I’m also focusing on here. In part based on fit with phenomenology, she defends non-naturalist realism about normative reasons. She argues that the alternative, a naturalist view, can’t account for the independent authority of the experienced demands.

The reasons-involving account, independently of how it’s developed ontologically, is vulnerable to an important objection, however (e.g., Dancy, 2006). The objection rests on a distinction between contributory normative reasons and the reasons-relative ought, where the latter depends on what there is most reason to do. On Hampton’s view, the experience of a demand with independent authority can be explicated as the apprehension of reasons.¹⁷ But, the objection goes, reasons are merely contributory, and, as such, they do not carry the sort of authority that our moral phenomenology suggests. Contributory reasons are typically understood as considerations that favor a certain action (Scanlon, 1998). They do not entail an ought. An ought, in the sense of an obligation, is an overall notion and it is binding. Normative reasons, by contrast, are contributory and they do not bind. So the attempt to explain the phenomenology of direct moral demands as an experience of (contributory) reasons fails.

Of course, this is not to say that we can’t account for obligations on the basis of reasons. On a common view, which I share, moral obligations are determined by what there is most reason to do.¹⁸ What there is most reason to do depends on a weighing of reasons, however. And the weighing of reasons depends on some form of deliberation or reasoning. If that’s correct, then we can’t account for the experience of direct moral demands in terms of an immediate apprehension of what there is most reason to do. As explained above, while moral deliberation, and deliberative moral judgments, are important elements of moral phenomenology overall, the phenomenology of direct moral demands is non-deliberative.

To illustrate the issue in the pond example, while there is a – weighty – normative reason to rescue the child, that reason, on its own, can’t properly accommodate the experience of a demand with independent authority. And that’s because your obligation to rescue the child is constituted by what you have overall reason to do. What does support the judgment that you ought to rescue the child is that you have a decisive reason to rescue the child – competing reasons are much less weighty than the reason to rescue the child. But a judgment about a decisive reason is deliberation-dependent – it requires appropriately weighing relevant contributory reasons and reaching a conclusion about what you have most reason to do in the circumstances. Because such a judgment is deliberation-dependent, it is in tension with the experience of direct moral judgments.

I thus conclude that neither contributory reasons nor the reasons-involving ought are well-suited to account for the experience of a demand with independent authority. Reasons are, I believe, very important in other normative contexts, both in ethics and epistemology. In ethics, they are importantly involved in the explanation of moral obligation, as well as in moral deliberation about our own actions and those of others and in moral justification. But they don’t help us account for the experience that certain situations make moral demands on us.

Can we can account for the experience of direct moral demands in terms of moral obligations directly? Moral obligations are either normatively basic or constituted by other normative facts, for example by normative reasons. As I’ve already argued against an account involving a non-basic ought, i.e., the ought constituted by normative reasons, I’ll now focus on ought understood as a basic normative property.

The obligations-involving account of the phenomenology of direct moral judgments looks much more promising than the reasons-involving account. Moral obligations make demands on us that are binding independently of our desires or similar attitudes. They thus appear to be a natural match for the experience that there is a demand with independent authority. This said, while an obligations-involving account of the moral phenomenology under consideration has the advantage of being normatively quite straightforward, it’s worth flagging the possibility that it might be too strong. An obligations-involving account excludes the possibility that supererogatory actions – actions that are in some sense appropriate but not obligatory – are experienced as being demanded by the circumstances. I think that sort of experience is not uncommon – not all demands are obligatory, not even all moral demands. I’ll come back to this point in the next section.

The main challenge for an obligations-involving account is metaphysical, however, not normative. What is the ontological status of moral obligations, if we take them as normatively basic properties? The options on the table are naturalism, either realist or anti-realist, and non-naturalist realism. On a non-naturalist view, such as the one pioneered by Samuel Clarke, for example, such obligations, or duties, are part of the fabric of a distinct moral reality. They are normatively and metaphysically basic and can’t be explained on the basis of natural properties.¹⁹ That sort of realist view has been the target of J.L. Mackie’s influential critique (Mackie, 1977). As a result of this critique, many philosophers, myself included, take it that one of the important tasks of meta-ethics is to explain the existence of moral obligations in a way that avoids overly demanding metaphysical commitments (see e.g., Darwall, 2006; Schroeder, 2007; Wallace, 2019). Given non-entailment, and in the interest of avoiding an overly demanding meta-ethical view, metaphysical primitivism about moral obligations thus does not seem promising.

But perhaps we can account for the moral phenomenology under consideration if we rely on an anti-realist meta-ethical view? Horgan and Timmons (2018) take this route. They argue that their preferred meta-ethical view – they call it cognitivist expressivism – can accommodate the experience of moral demands with independent authority without presupposing non-naturalist realism about ought. Their sophisticated view is cognitivist because it construes moral judgments as involving beliefs, specifically beliefs tracking “ought-commitments” (2018: 327). But it’s still a form of anti-realism because ought-commitments are a kind of psychological commitment and beliefs about ought thus do not track an objective moral reality.²⁰ Their proposal is that the felt moral demand is a belief about what you ought to do and that belief tracks your ought-commitments.

While their obligations-involving proposal avoids the demandingness objection, it runs into a different objection, however. The objection is that cognitivist expressivism can’t fully accommodate the moral phenomenology under consideration. Specifically, because it construes moral judgments as contingent on ought-commitments, it can’t fully accommodate the independent authority of the felt moral demand. The source of a felt demand, on this account, is an ought commitment that we have made, not something outside of us (FitzPatrick, 2021, p. 10).²¹

Note that this objection, which we might call the no-authority objection, is compatible with accepting Horgan and Timmons’ non-entailment claim. The objection is not that the phenomenology of direct moral demands entails a realist metaphysics. The objection is, rather, that cognitivist expressivism fails to do full justice to the moral phenomenology under consideration because the experience of independent authority is not accommodated. Instead of accounting for the experience of a demand with independent authority, that experience is explained away as a kind of illusion.

Because Horgan and Timmons offer a sophisticated version of anti-realism about moral obligations, we should expect the no-authority objection to generalize to other anti-realist views of moral obligations. If non-naturalist realism and anti-realism about ought are ruled out, however, this only leaves naturalist realism about moral obligations as a final option. On that kind of view, there are true moral judgments involving ought, but the obligations that are involved in such judgments can be explained on the basis of natural properties – for example via the cultural evolution of moral codes. What should we think about that kind of view?

An important objection is that because this sort of view rests on social or psychological contingencies, it is vulnerable to a version of the no-authority objection. That is to say, because obligations are relative to some psychological or social processes, this kind of view can’t appropriately accommodate the independent authority of the felt demands (Horgan & Timmons, 2018, p. 322, Hampton 1998). What’s being highlighted is a tension between the non-contingent bindingness of moral obligations, on the one hand, and the contingent social or psychological origin of those obligations, on the other. This objection thus also rules out the last option for accounting for the phenomenology of direct moral demands in terms of moral obligations.²²

In sum, I’ve argued that normative reasons are not well-suited to accommodate the experience of direct moral demands because such reasons are contributory and not-binding. Obligations are more promising in this regard, but plausible obligations-involving accounts either create metaphysical difficulties or they struggle with accommodating the independent authority of moral demands.

6. Accounting for the phenomenology of direct moral demands

In this final section of the paper, I explain how the fittingness-involving account of the phenomenology of direct moral demands that I developed earlier is superior to the main alternatives. I defend three main claims in this section. First, moral affordances, understood as opportunities for fitting action, shed fresh light on the kind of demand that is being experienced. Second, moral affordances can account for the independent authority of those demands. And, finally, moral affordances help us get a grip on the experience of a demand with independent authority – through the possibility that moral affordances are being experienced as solicited by the environment.

Regarding the first, we saw in the previous section that existing proposals for how to account for the phenomenology of direct moral demands involve either reasons or obligations. Reasons, I’ve argued, do not offer us a good account of the experience of direct moral demands because they are contributory. Obligations do better in this regard. But, as we saw, the ontology of obligations that are not derived from reasons creates its own difficulties for an account of the phenomenology of direct moral demands, facing either a demandingness objection or a no-authority objection.

The fittingness-involving account that I’m proposing, by contrast, fares better. Like obligations, fittingness, as I understand it, is an overall notion, not merely a contributory notion such as reasons. A fitting response, as I’ve explained above, is the response that is merited by a given situation. That makes fittingness an overall notion. It picks out how to respond to a situation, it doesn’t merely contribute to an overall assessment of how to respond. As an overall notion, fittingness entails a kind of ought, and that’s why a fittingness-involving account is more suited to the phenomenology of direct moral demands than a reasons-involving account.

The ought of fittingness, although an overall notion, is not binding, however, unlike the ought of moral obligation.²³ We might say that fittingness is orientational for moral action, in the same way that truth is orientational for belief. Bindingness is a key feature of the ought of moral obligation, whether understood as normatively primitive, or, as I prefer, as constituted by what there is most reason to do. If there is a moral obligation to do x, doing x is morally required and there are justified sanctions if the obligation is not met.²⁴ A fitting response, while being the response that is appropriate or warranted in given circumstances, isn’t required in the way moral obligations are and a failure to respond accordingly need not justify sanctions. There is an important difference between the response that a situation demands and the response that can be required of an agent. That difference is the normative difference between the ought of fittingness and the ought of moral obligation. Because the ought of fittingness is weaker than the ought of moral obligations, fittingness creates fewer ontological challenges than obligations.

The normative difference between fittingness and obligation is easy to overlook. One reason for this is that, in many cases, the ought of fittingness and the ought of moral obligation pick out the same action. In the case of the child that has fallen into a pond right in front of you, rescuing the child is fitting and you also have a moral obligation to save the child. We can explicate that obligation in terms of what you have most reason to do in this case.²⁵

But in other cases, the two oughts can come apart and they can even pull in different directions. They come apart in some attitudinal cases. For example, while it’s fitting, say, to be sad after the death of a loved one, there is normally no obligation to be sad. As I would explain this, there are reasons to be sad, but those reasons aren’t such that not being sad would be morally impermissible.²⁶ Fittingness and obligations can also come apart in relation to action. In some cases of supererogatory action, for example, while there isn’t an obligation to perform that action, it will be fitting to do so. Because it’s fitting, there is a sense in which you ought to perform that action, but the ought isn’t the ought of obligations.

Fittingness and obligations can also pull in different directions, such that what’s fitting to do conflicts with what there is an obligation to do. When they do, the ought of moral obligation normatively trumps the ought of fittingness because the former but not the latter is binding. An obligation incurred through a promise, for example, might be in tension with what would be a fitting response to the situation. But that tension doesn’t nullify the obligation if the fitting response isn’t itself obligatory.

If fittingness entails a kind of ought, as I’ve argued, this explains why moral affordances, understood as opportunities for fitting action, shed fresh light on the kind of ought that figures in experiences of direct moral demands. The moral affordance of a given situation is the appropriate, or merited, response to this situation. It doesn’t merely contribute to the assessment of what the right response is, as reasons do, but nor does it entail that the response is obligatory. Moral affordances contain a distinct type of moral demand.

What about the second feature of the phenomenology of direct moral demands, that the demands are being experienced as having an independent authority? Moral affordances are also well-suited to accommodate this feature, or so I want to argue now. As we just saw, accounts involving either fittingness or obligations are better suited than reasons-involving accounts to cover the demand aspect. But the fittingness-involving account that I’m developing in this paper has advantages over the obligations-involving account with regard to the authority aspect. And the reason lies in the ontological advantages of my fittingness-involving account compared to an obligation-involving account.

Before I proceed, let me briefly address an objection against my account. The objection is that a fittingness-involving account runs into a version of the no-authority objection. And the explanation for this is that fittingness can’t explain the experience of a demand with independent authority of moral demands because fittingness is response-dependent. What makes a response fitting are features of the response, not features of the world (e.g., D’Arms & Jacobson, 2000).

My reply to this objection draws on the external standard view of fittingness that I introduced in section 4. The view that fittingness is response-dependent assumes an internal standard view, according to which fittingness depends on features internal to a response. In addition, it assumes that the standard of correctness for fitting responses is set by the responder, whether it’s an idealized responder or an actual responder. As you might recall, I rejected an internal standard view of fittingness. On the external standard view that I defended, what makes a response fitting depends on features of the situation as a whole. Moral affordances incorporate this external standard view. They can be described by a three-place relation between an agent, the circumstances they find themselves in, and the agent’s fitting response to their circumstances. While moral affordances depend on features of the agents, they are not purely subjective. As such, moral affordances are well-suited to capture the objectivity of direct moral demands, the feature that a certain response is called for by the circumstances. They fair better in this regard than naturalist accounts of moral obligations that end up being overly subjectivist and that run into a version of the no-authority objection for that reason (see section 5).

But moral affordances, as I understand them, also have an ontological advantage over non-naturalist versions of an obligations-involving account. As we saw above, many philosophers doubt the existence of metaphysically (not just normatively) primitive moral obligations. My proposal is less demanding for two reasons. First, moral affordances do not make binding demands, unlike moral obligations. The demands of moral affordances are thus easier to explain. Second, moral affordances are at least in principle compatible with a naturalist view of what makes actions fitting. Admittedly, in this paper, I’ve focused on the formal features of an account of the phenomenology of direct moral demands. I’ve left open what makes actions fitting, substantively speaking. But we can note that nothing in my proposal commits us to non-naturalism about moral demands. While I don’t have the space to develop a naturalist interpretation of fittingness here, the lack of a commitment to non-naturalism is a further reason to favor moral affordances over alternative attempts to account for the phenomenology of direct moral demands.

A final feature of my moral affordances proposal that is worth emphasizing is that it has the potential to shed light on the experience of direct moral demands. As I explained earlier, a key claim of affordance theory is that affordances are directly perceivable. And since moral affordances, as I understand them, are a subset of affordances, we should expect moral affordances to be perceivable as well. Of course, this is not to say that moral affordances are always perceived. It is possible to overlook moral affordances, just as it’s possible to overlook other affordances of the environment. Moreover, it’s also possible, I take it, to perceive moral affordances and not be moved by them. And that is possible for two reasons. First, it’s possible that a moral affordance is perceived, but it’s not perceived as a moral affordance. So it’s only perceived as a possible response in the given circumstances, but it’s not perceived as the fitting response. Alternatively, it’s also possible that while an affordance is perceived as the fitting response, it fails to motivate action because self-interested motivations dominate.

Because this paper focused on the content of moral experiences, the explanation of how moral affordances might be perceived has to be given elsewhere. But let me note that an affordances account is in principle well-suited to spell out the experience of moral demands as a type of perceptual experience. What is more, assuming that moral affordances are also the sort of affordances that can be perceived as being mandated by the environment, we then we have a full account of the phenomenology of moral demands. On this account, moral affordances are perceivable as the response that is being called for by the circumstances.

7. Concluding remarks

In this paper, I’ve argued in favor of a fittingness-involving account of some paradigmatic moral experiences – the experience of direct moral demands. The account focused on moral affordances understood as opportunities for fitting action. While much work remains to be done, I’ve shown that affordance theory lends itself to further our understanding of moral phenomenology and of the nature of moral demands.

Acknowledgements

I did the research for this paper while I was a Faculty Fellow at the Murphy Institute at Tulane University in New Orleans. I’m very grateful to the Institute for providing a wonderful research environment. I received helpful feedback on this paper at a workshop on “Moral Experience” that Daniel Vanello and I co-organised at the University of Warwick in March 2023. I particularly want to thank Sophie-Grace Chappell, Robert Cowan, Naomi Eilan, William Fitzpatrick, Guy Longworth, Eileen John, Max Khan Hayward, and Daniel Vanello. I also want to thank two anonymous referees for their very constructive comments.

Disclosure statement

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

Notes

1. Exceptions include Jayawickreme and Chemero (2008), Wisnewski (2019), Hampson et al. (2021), Rottschaefer (2021), and van den Herik and Rietveld (2021).

2. See FitzPatrick (2018) for a distinction between two types of meta-ethical inquiry, one starting from moral experiences and the other from metaphysical or epistemological presuppositions.

3. I benefitted greatly from Horgan and Timmons’ work on Mandelbaum’s moral phenomenology, especially Horgan and Timmons (2007, 2008, 2017, 2018). See also Hampton (1998) for an examination of related aspects of moral phenomenology.

4. Mandelbaum (1955) distinguishes direct moral judgments from removed moral judgments and judgments of the moral worth of actions.

5. See Horgan and Timmons (2018, p. 326).

6. There is an extensive debate on the ontology of affordances, not least in an attempt to clarify the idea that affordances are neither objective nor subjective. While there is some disagreement on this issue, the dominant view in the literature is that affordances are relations. The main alternative to the relational view is a dispositional view, according to which affordances are dispositions of environmental objects (Turvey, 1992, Scarantino 2003). Vetter (2020, p. 1188) argues that the debate on the ontology of affordances may be overstating the issue as, she claims, her preferred version of a dispositional theory can accommodate the relational view of affordances. I won’t take a stand on whether it can or not, but simply assume that affordances are relational, which she also accepts.

7. See the quote from Gibson (2014) above.

8. This is loosely based on Chemero (2003, p. 189–191).

9. “The environment of any animal … contains substances, surfaces and their layout, enclosures, objects, places, events, and the other animals” (Gibson 2014: 31).

10. Note that while Siegel (2014) is sympathetic to the phenomenology of direct perception, she explores the possibility of a representational account.

11. While I’m bracketing it here, I’ll discuss the comparison between fittingness, reasons, and obligations in section 6.

12. On some prominent views, fittingness isn’t a basic normative property or relation, but can be explained in terms of reasons (Schroeder, 2021). But on an increasingly popular view, fittingness is a basic normative property or relation in its own right (e.g., Howard, 2019; McHugh & Way, 2022). In the next section, I argue that a reason-based approach isn’t helpful to account for the phenomenology of direct moral demands. In this section, I want to leave reasons-based views to the side and explore the role of fittingness, understood as a basic normative relation, in moral affordances.

13. This characterization echoes Howard’s (2018) “gloss” of fittingness, but broadens it to focus not just on objects, but to a situation as a whole. The understanding of fittingness as an appropriate or merited response has a long history in moral philosophy. It can be found in Aristotle’s Nicomachean Ethics and Brentano offered an analysis of goodness in terms of what is fitting to love (Brentano, 2009).

14. Exceptions include Chappell (2012) and Howard and Leary (2022).

15. D’Arms and Jacobson (2000) argue that this sort of view commits a moralistic fallacy, as it conflates the fittingness of envy, which they interpret as relative to the type of response, and a moral judgment of envy. I accept that there is a distinction between normative features of types of responses and moral features of responses. But I’m interested in the latter here, and I think fittingness is useful to shed light on that.

16. Note that the properties or relations that are involved must be normative, not merely evaluative, and that rules out values. The experience isn’t merely that it would be good to respond in a particular way. Instead, as we saw, the experience is of a demand.

17. On the perception of reasons, see Church (2010).

18. See Schroeder (2021) for a recent articulation of this, now fairly common, view of the relation between normative reasons and obligation.

19. See e.g., Korsgaard (1996: 28f) for a discussion.

20. “According to our non-reductive metaethical expressivism, a moral judgment … is a psychological commitment-state with respect to a potential way the world might be – an ought-commitment” (Horgan & Timmons, 2018, p. 327).

21. FitzPatrick (2021, p. 10) puts the objection as follows: “Since expressivism claims that in fact all normative governance ultimately obtains contingently … , it therefore must take this aspect of the phenomenology to be misleading: the appearance is that the basic standards are (somehow) imposed on us independently of our contingent psychologies, and expressivism insists that this is not actually what is happening, which is instead purely sociological and psychological.”

22. As will become clearer in the next section, however, not all naturalist accounts of the moral phenomenology under consideration are equally vulnerable to the no-authority objection.

23. The idea that there are two kinds of ought – the ought of fittingness and the ought of moral obligation – is consistent with Ewing’s understanding of fittingness. Here’s how Ewing puts the point (1939: 3): “In the one sense ‘the act I ought to perform’ stands for the act which is most fitting … in view of the situation; in the other sense it stands for an act neglect to perform which would be morally bad.” On Ewing’s view, whereas the ought of fittingness is normatively and metaphysically primitive, the ought of moral obligation can be analyzed in terms of fittingness and further considerations. Fittingness captures what a situation demands of a moral agent. For there to be a moral obligation, by contrast, further conditions must be met. See also Hurka and Hurka (2014) and Darwall (2017) for discussions of this distinction.

24. See Wallace (2019: 26ff).

25. This sort of – fairly standard – case raises interesting questions about the relation between fittingness, reasons, and obligations. This has to be addressed elsewhere.

26. I’m not arguing that there can’t be moral obligations for attitudes. I think there can. For example, there can be an obligation to repent. See Maguire (2018) for a different account. Maguire’s account rests on a sharp distinction between attitudes and actions, which I reject. See section 5.