The Object of Moral Understanding

1. The Object Question

My aim in this paper is to make progress on the question of what is understood in moral understanding. Call this the Object Question, since it asks what the cognitive objects of moral understanding are. The Object Question is not explicitly addressed in the literature on moral understanding.¹ Nevertheless, most recent work on moral understanding seems to converge on a common view of the object of moral understanding. This is because it is commonly assumed that moral understanding is a species of understanding why p. As Alison Hills puts it: “moral understanding [is] understanding why p (where p is some explicitly moral proposition).”² Paradigmatic instances of moral understanding, on this picture, include understanding why eating meat is wrong,³ understanding why plagiarism is wrong,⁴ and understanding why some particular action is the right thing to do.⁵

Call the view that moral understanding is a species of understanding why p the ‘explanatory picture’ of moral understanding. The explanatory picture of moral understanding provides a straightforward answer to the Object Question: what is understood, in each instance of moral understanding, is why some explicitly moral proposition is true.⁶

I will argue that the explanatory picture of moral understanding is incorrect, or at least incomplete, and contrast it with a different answer to the Object Question that highlights the role of moral understanding in reasoning about what to do. In many important instances of moral understanding, I show that what is understood is not why a moral proposition is true, but rather how a consideration bears on what to do. When someone understands how a consideration bears on what to do, the object of their understanding is the relation of normative support between a proposition and an action.

My attempt to make progress on the Object Question will thus have a negative and a positive component. First, I discuss two problems for the view that all instances of moral understanding are instances of understanding why some moral proposition is true. The first problem concerns agents who possess explanatory understanding of moral facts, but don’t understand how those facts bear on what to do (§2). The second problem concerns agents who manifest partial moral understanding but lack explanatory understanding (§3). Finally, in the last section of this paper (§4), I sketch a different answer to the Object Question that avoids these two problems, and argue that that it is plausible on independent grounds as well.

2. Explanatory Understanding and Practical Deliberation

The fact that some course of action would be wrong is relevant to what I should to do, all things considered. In particular, the wrongness of an action often provides decisive reason against doing it. Thus, suppose A realizes that it would be wrong to do X, and treats this consideration as a decisive reason not to do X. Here we can say that A understands how the wrongness of doing X bears on what they should do. This kind of understanding—understanding how the wrongness of an action bears on what one should do—is one of the most basic forms of moral understanding. Yet it is not accounted for by the explanatory picture. Or so I shall argue in this section.

The problem, to put it in the most general terms, is that one can understand why p is the case without understanding how the fact that p bears on what to do. One can be able to work backwards from the fact that p to the explanation of why p obtains, while failing to grasp the implications of p for the forward-looking question of what to do. For instance, suppose Nutro the selfish nutritionist understands full well why his partner is ill: he knows that she is ill because she ate undercooked chicken, and he can explain in great detail how eating undercooked chicken caused her to be nauseous. Yet Nutro fails to understand how the fact that she is ill bears on what he should do: he doesn’t treat the fact that she is ill as a reason to take care of her. He understands why p, but not how the fact that p bears on what to do.

Now, the fact that someone is feeling ill is not itself a moral fact, so this example does not yet raise any difficulties for the explanatory picture of moral understanding. But it provides the template for such an example. Just as the fact that someone is ill provides a normative reason to try to relieve their discomfort, so too the fact that a course of action would be morally wrong provides a strong normative reason not to act in that way. The distinction between understanding why p and understanding how p bears on what to do thus raises the following possibility: one might be able to work backwards from the fact that an action is wrong to the reasons why it is wrong, while failing to grasp the implications of that fact for what one should to do.

To illustrate, imagine a ruthless hedge fund manager who is considering buying the rights to a medication many people desperately need, and hiking the price of that medication so as to make a massive profit. On the one hand, our hedge fund manager realizes that doing this would be morally wrong and why it would be wrong: hiking the price of the medication would be wrong, he acknowledges, because it would be unjustifiable to all those who could no longer afford the medication (and even to those who still could), and it would be unjustifiable to these people because they could reasonably reject any principle that allows someone’s predatory profiting at the expense of their health.⁷ So our hedge fund manager understands why it would be morally wrong to buy the rights to the medication and make it excessively costly. Nonetheless, he doesn’t treat the wrongness of that action as a reason not to do it. ‘So what’, he says to himself, ‘if I can’t justify my actions to those affected by it. Who says I have to?’ Instead, the hedge fund manager believes that he should do whatever maximizes profits. This goal will be achieved by acquiring the rights to the medication and price gouging. So, that is what the hedge fund manager believes he should do. And that is exactly what he does.⁸

The hedge fund manager is mistaken about what he should do. That is, he is mistaken about which course of action is best supported by the normative reasons. Price gouging is not, in fact, what he should do. The hedge fund manager arrives at this mistaken belief about what he should because he disregards what is in fact a weighty normative reason: that price gouging would be wrong.⁹

The hedge fund manager’s failure to treat this fact as a reason is a failure of moral understanding. It is a failure to appreciate what follows from the fact that price gouging would be wrong. In other words, it is a failure to understand something about a moral fact: its reason-giving force. And yet, by stipulation, the hedge fund manager’s explanatory moral understanding is impeccable. He understands full well why price gouging would be wrong.¹⁰ Thus, there is more to moral understanding than merely understanding why certain moral propositions are true.

Proponents of the explanatory view of understanding might object that I am conflating two different kinds of shortcomings: cognitive mistakes, and failures of motivations. After all, as I have presented him, the hedge fund manager is a paradigm example of the philosopher’s amoralist: someone who realizes that a course of action would be wrong, but who just doesn’t care. This is clearly a moral failing, but it is much less clear that it is a failure of moral understanding. After all, understanding is a cognitive or intellectual achievement. The hedge fund manager seems to be doing just fine on the cognitive side of things: he understands why price gouging would be wrong. This suggests a different verdict about the case: the hedge fund manager has an impeccable moral understanding, but bad motivations.

I agree that the hedge fund manager is not properly motivated. But I don’t think that this criticism competes with my criticism of his moral understanding. Rather, I think that both forms of criticism are appropriate in our case: the hedge fund manager’s moral understanding is flawed and so too are his motivations.

To support this interpretation, consider a variation of our case in which our character’s only failing is motivational. In this different version of the case, the hedge fund manager realizes that it would be wrong to price gouge and, on that basis, comes to the conclusion that he should not acquire the rights to the medication and hike the price. But, for whatever reason, the hedge fund manager cannot bring himself to be motivated in line with his belief about what he should do. Perhaps the lure of a massive profit is too hard to resist and the hedge fund manager finds himself acting against his better judgment. Or perhaps the hedge fund manager is in the midst of a devastating break-up, and finds himself in the self-destructive mood of being motivated to do precisely what he believes he ought not to do.¹¹

Note that that there is an important cognitive difference between the agent in this version of the case and the one with which we began. The weak-willed or self-destructive versions of the character holds the right cognitive states: they correctly grasp the implications of the fact that price gouging would be wrong for what they should do. They draw the right inference from one belief (‘price gouging would be wrong’) to another belief (‘I shouldn’t price gouge’). Their shortcoming is purely motivational. They are unable to bring their motivations into line with their belief about what they ought to do.¹² By contrast, the hedge fund manager in our original version of the case doesn’t draw this inference at all. He not only fails to be motivated by the fact that price gouging would be wrong, but also fails to realize that he shouldn’t price gouge. Unlike the weak-willed or self-destructive versions of the character, he fails to understand how the wrongness of price-gouging bears on the further question of what he should do. His failing is, in part, a failure of moral understanding.

The problem for the explanatory picture thus stands. It fails to account for a very basic form of moral understanding: understanding how the wrongness of an action bears on what I should do.

Part of what makes this form of understanding so basic or fundamental is that it is arguably part of the nature of moral wrongness that the wrongness of an action provides normative reasons against it. This observation, in turn, suggests a different reply on behalf of the explanationist: how could someone understand why an action is wrong when they don’t fully understand what it is for an action to be wrong? The thought here is that you can’t understand why it is the case that p if you don’t understand the nature of the fact that p. Drawing on this principle, the explanationist might deny that the hedge fund manager really does understand why price gouging would be wrong. By failing to treat the wrongness of price gouging as bearing on what they should do, the hedge fund manager reveals that he doesn’t really understand what it is for that action to be wrong. Since he doesn’t understand what it is for that action to be wrong, he couldn’t possibly understand why that action is wrong. Thus, the explanationist can easily accommodate our verdict that the hedge fund manager has a deficient moral understanding: he lacks explanatory moral understanding.

This objection stands or falls with what we might call the ‘essentialist’ requirement on explanatory understanding. According to this principle, understanding why x is F requires understanding what it is to be F. That is, one must understand the nature of the properties contained in the fact that P in order to understand why it is the case that P.¹³

The essentialist requirement is overly strong. Suppose Noor believes that ‘bachelor’ refers to someone who is either an unmarried man or a man who hasn’t talked to their spouse in more than a month. Noor does not understand the nature of bachelorhood: she thinks the property of bachelorhood encompasses far more than it in fact does. Still, Noor can understand why her brother, who is irascible, selfish, and plays video games all day, is a bachelor. She understands why her brother is a bachelor because she understands why he is unmarried, and she correctly takes this to account for his being a bachelor.

The reason why the essentialist requirement is overly strong is that it does not distinguish between the features of an explanandum that are relevant to a particular explanation and those that are not. To illustrate, consider a different example. It is part of the nature of ophthalmologists that if someone is an ophthalmologist, then that person is a doctor. Can Alice understand why Zoe became an ophthalmologist, even if she does not know that ophthalmologists are doctors? The answer is that it depends on whether the fact that ophthalmologist are doctors is explanatorily relevant to why Zoe became an ophthalmologist. If Zoe became an ophthalmologist because the job pays well and allows her to help other people, and Alice knows that ophthalmologists are well paid and help other people, then it seems that Alice is a good position to understand why Zoe became an ophthalmologist. Even though Alice has an imperfect understanding of what an ophthalmologist is, she can give a perfectly good explanation of why Zoe became an ophthalmologist. If, on the other hand, Zoe became an ophthalmologist simply for the prestige of being a doctor, and Alice does not know that ophthalmologists are doctors, then Alice is not in a good position to understand why Zoe is an ophthalmologist. The key difference between the two variants of the case is that, in the second, the fact that ophthalmologists are doctors is explanatorily relevant: it explains why Zoe became an ophthalmologist.

The crucial point, for our purposes, is that the case of the hedge fund manager is analogous to the first variant of Alice’s case. The fact that the wrongness of an action provides decisive reason not to perform it is not explanatorily relevant to why the action is wrong. Hiking the price of the medication is wrong because it is unjustifiable to those who could no longer afford the medication, and it would be unjustifiable to these people because they could reasonably reject any principle that allows someone’s predatory profiting at the expense of their health. This explanation does not refer to the fact that the wrongness of the action is a decisive reason against it. That is not surprising. The distinctive reason provided by the wrongness of an action cannot be what explains why it is wrong in the first place.

Thus, even if we grant that the hedge fund manager has an imperfect understanding of the nature of wrongness, this still wouldn’t show that he doesn’t understand why hiking the price of the medication is wrong. The hedge fund manager, after all, has some understanding of what wrongness is: he thinks that the wrongness of an action consists in its unjustifiability to others. His thoughts involving ‘wrongness’ thus plausibly succeed in referring to wrongness. And while he is in one important respect mistaken about what wrongness is, that mistake is not about some feature that explains why his action is wrong. Because the mistake is not about some feature that explains why his action is wrong, it does not prevent him from understanding why hiking the price of the medication is wrong.

To summarize: there are at least two ways to manifest moral understanding that is about the wrongness of an action. One is by understanding why that action is wrong. Another is by understanding how the wrongness of that action bears on the further question of what I should do. These forms of moral understanding are independent. One might understand full well why an action is wrong without understanding how that fact bears on what one should do. Thus, there is more to moral understanding than merely understanding why certain moral facts are true.

The argument I have explored in this section has the following structure: it attempts to isolate what the explanatory picture of moral understanding leaves out by considering moral agents whose explanatory moral understanding is impeccable, but whose moral understanding seems deficient in other ways. In the next section, I explore an argument that has the reverse structure: I attempt to isolate what the explanatory picture of moral understanding leaves out by considering moral agents who lack explanatory moral understanding, but whose moral understanding nevertheless seems good in other ways.

3. Factivity and Partial Moral Understanding

Explanatory understanding is factive with respect to its explanandum. S understands why p only if it is true that p. Given that Argentina won the 1986 World Cup, it is impossible for anyone to understand why Brazil won the World Cup that year. This factivity condition on what is explained holds for both full and partial understanding why. It’s not just that no one can fully understand why Brazil won the 1986 World Cup; no one can partially understand why Brazil won the 1986 World Cup, either, given that they didn’t win. In short, for any p, S understands (fully or partially) why p only if it is the case that p.¹⁴

This feature of understanding why gives rise to a problem for the explanatory picture of moral understanding. In some hard cases, people manifest a good but imperfect understanding of the morally relevant features of their situation. Because their moral understanding is imperfect, or partial, they draw the wrong conclusion about what the right thing to do is. Since they are mistaken about what the right thing to do is, they are not in a position to understand why the action that is the right thing to do is the right thing to do. They exhibit partial moral understanding, but lack explanatory understanding.

To illustrate, suppose Anna is a journalist preparing to go on a reporting trip to a dangerous and conflict-ridden area. She has to tell her family that she will be away but she doesn’t know how much to tell them. She knows that, if she tells them exactly where she’s going and why, her parents will spend sleepless nights worrying about her safety. On the other hand, she knows that she can’t keep the truth from them without lying. Anna thinks through her predicament carefully and comes to the conclusion that the right thing to do is to keep her family in the dark, even if that means lying to them. However, let us stipulate that Anna is mistaken about this. In this particular case, the right thing for her to do is to tell her family the truth.¹⁵ Call this case Trip.

If telling the truth is the right thing to do, then keeping her family in the dark isn’t. So Anna cannot understand, partially or in full, why keeping her family in the dark is the right thing to do. Nor does Anna understand why telling her family the truth is the right thing to do. This is because she fails to satisfy a basic cognitive condition for explanatory understanding with that content, namely: in order to understand why p, you have to believe that p. Anna doesn’t even believe that telling her family the truth is the right thing to do. A fortiori, she doesn’t understand why telling her family the truth is the right thing to do.¹⁶

Here is where we seem to have ended up. Because Anna draws the wrong conclusion about what is, all things considered, the right thing to do, she lacks explanatory moral understanding of what the right thing to do is. The explanatory picture of moral understanding thus yields the verdict that Anna lacks even a partial understanding of her situation. But this seems overly harsh. Anna is sensitive to those features of her choice that are in fact morally salient. She treats the fact that evading her family’s questions would amount to lying, correctly, as a reason not to keep them in the dark. And she treats the fact that they would worry if they knew the truth, correctly, as a reason not to tell the truth. Although she draws the wrong conclusion about what to do, she clearly has a partial, if not perfect, moral understanding of her situation.¹⁷

Anna is thus a paradigm example of someone who has a partial understanding of the morally relevant features of her situation, but her moral understanding is not explanatory. This is bad news for the explanatory account, and one that generalizes to any hard case in which an agent has a reasonably good understanding of their moral reasons yet draws the wrong conclusion about what to do.

Proponents of the explanatory model might acknowledge that Anna does have partial moral understanding, and concede that this partial moral understanding cannot be accounted for in terms of her understanding why φ-ing is the right thing to do. But, they will insist, this does not mean that Anna’s moral understanding is not explanatory. It just means that Anna’s moral understanding consists in her understanding why some other moral fact obtains.

For instance, the proponent of the explanatory picture might suggest the following: what Anna understands is why causing others emotional pain is typically or normally wrong. However, this does not seem to me to capture the moral understanding that Anna manifests in this particular case. To bring this point home, consider Xavier, a photojournalist working on the same team as Anna. Like Anna, Xavier has to decide whether to tell his family about the dangerous location where he will be working over the next few months. And let us suppose, further, that Xavier understands why it’s typically wrong to cause others emotional anxiety. But, for whatever reason, the fact that his family will be worried sick about his safety if he tells them where he is going doesn’t weigh with him at all in deciding whether to tell them where he is going. Perhaps he is so excited about the opportunity to do photojournalism abroad that it hasn’t even occurred to him that this might affect others adversely. Or perhaps he resents the helicopter parenting he had to endure as a child, and is disposed, as a result, to ignore his parents’ concerns about his wellbeing. Either way, Xavier fails to appreciate how the emotional anxiety he will cause his family if he tells them where he is going bears on what he ought to do in this case. Even though he understands why it is typically wrong to cause others emotional anxiety, he lacks the moral understanding that Anna achieves. Thus, the moral understanding that Anna manifests isn’t just an understanding of why a certain act type is typically right or wrong.

A more promising avenue for the explanationist is to focus on moral features of Anna’s particular circumstances of which she might have explanatory understanding. For instance, it might plausibly be suggested that even though Anna does not understand why lying to her family is wrong all things considered, she may understand why lying to her family is pro tanto wrong, or morally problematic. After all, one can believe that an action is morally problematic in some respect without believing that it is wrong all things considered. In this way, the explanationist seems able to accommodate our intuition that Anna has a good if imperfect moral understanding of her circumstances. While Anna may not understand why lying is wrong all things considered (since she doesn’t believe it is), she may nevertheless understand why it is wrong to some extent.

More generally, the basic strategy is to distinguish between what we might call ‘intermediate’ and ‘conclusive’ moral beliefs, and to imagine moral agents in hard cases as forming intermediate moral beliefs (e.g., that an action is pro tanto wrong) on their way to coming to a conclusive moral verdict (e.g., that an action is the right thing to do). Since moral agents can get things right at the intermediate stage of moral reasoning while being mistaken about what is the right thing to do all things considered, the explanatory account of moral understanding has more purchase at this intermediate level. Thus, whenever confronted by cases in which moral agents draw the wrong conclusion about what to do, yet seem to display a good moral understanding of their circumstances, explanationists should reply by drawing our attention to the explanatory understanding of intermediate moral facts which such agents display.

This strategy, I believe, runs into two problems. The first is that, in some cases, moral agents may be mistaken at the level of intermediate moral judgment as well, without this undermining our sense that they have a good moral understanding of their circumstances. The second is that it is conceivable that moral agents might not form any intermediate moral beliefs about their circumstances, again, without this undermining our sense that they have a good if imperfect moral understanding.

To illustrate these limitations of the explanationist’s reply, consider a different case: Sartre’s student. As Sartre tells the story, one of his students, call him ‘Pierre,’ finds himself in a difficult predicament during the German occupation of France. On the one hand, Pierre feels compelled to join the underground military resistance. On the other hand, Pierre’s mother lives all by herself and she is already grief-stricken (her son, Pierre’s older brother, has recently died in the resistance). There is no saying what will happen to her if Pierre, like his brother before him, joins the resistance too. Pierre finds himself genuinely torn about what the right thing to do is.¹⁸

Suppose that the morally right thing for Pierre to do is to stay with his mother. And suppose further that Pierre incorrectly concludes that the right thing to do is to instead join the resistance. So, Pierre doesn’t understand why the right thing to do is to stay with his mother (he doesn’t believe this), nor does he understand why joining the resistance is the right thing to do (it isn’t).

Pierre thus lacks explanatory moral understanding at the conclusive level. So far, so familiar. The explanationist grants this point. Their reply is to direct our attention to Pierre’s intermediate moral beliefs, and the explanatory moral understanding he displays there.

The first problem with this reply is that Pierre’s moral beliefs may be mistaken at the intermediate level, too. For instance, imagine that the two moral beliefs that Pierre forms, en route to deciding what to do, are the following. He believes that he would be a bad son if he abandoned his mother, and he believes that it would be dishonorable not to join the resistance while others are risking their lives to liberate France. These are the only intermediate moral beliefs Pierre comes to hold about his situation. And they are, let us stipulate, both mistaken. It would not be dishonorable for Pierre to stay with his mother, and he would not be a bad son if he joined the resistance. In applying these thick moral concepts, Pierre is being too hard on himself.

Since Pierre would not be a bad son for joining the resistance, nor acting dishonorably by failing to join, he cannot possibly understand why either of these two propositions is true. And yet, for all that, it still seems that Pierre possesses a good if imperfect understanding of his moral predicament. He is sensitive to what are in fact morally relevant features of his situation (his mother’s needs, the sacrifices made by his compatriots), even though the moral concepts in terms of which he thinks of those features do not strictly apply.

We can formulate the problem in more general terms. The explanationist’s strategy was to account for the partial moral understanding of moral agents who are mistaken about what the right thing to do by appealing to the explanatory understanding of intermediate moral facts which such agents allegedly possess. What our example illustrates is that whether agents actually possess such explanatory understanding at the intermediate level depends on which intermediate moral beliefs they actually hold. In some cases, the only moral beliefs which agents hold at the intermediate level may be mistaken as well.

The second problem with the explanationist’s strategy is an extension of the first. Just as moral agents can have partial moral understanding despite being mistaken in their intermediate moral beliefs, so too moral agents can have a good if partial moral understanding of their situation without holding any intermediate moral beliefs at all.

The conceptual possibility I have in mind is that of reasoning directly about what the right thing to do is, without forming any intermediate moral beliefs about one’s circumstances. For instance, suppose, now in a different version of the case, that the two options of joining the resistance and staying with his mother simply present themselves to Pierre with the force of being what he all-things-considered morally ought to do. When Pierre pictures his companions risking their lives in the resistance, it seems to him that joining the resistance must be the right thing to do. But when he pictures his mother, lonely and grieving in her small apartment, he feels with equal force that staying with his mother is the right thing to do. In trying to decide what the right thing to do is, Pierre is torn between two courses of action that present themselves with the force of all-things-considered moral obligation. Indeed, to make the case even more vivid, we can even stipulate that Pierre, in this version of the example, lacks any intermediate moral concepts. He can only reason, directly, about what is the morally right thing to do all things considered. And, as in the previous version of the case, Pierre forms the wrong belief about what is the right thing to do (believing, incorrectly, that the right thing to do is to join the resistance).

By stipulation, the only explicitly moral belief that Pierre holds about his situation is that he morally ought to join the resistance, and that belief, we are supposing, is false. Thus, there is no moral proposition about his situation such that Pierre understands why that moral proposition is true. Pierre lacks explanatory moral understanding. And yet it still seems too harsh to say that Pierre lacks moral understanding full stop. After all, Pierre is sensitive to the competing pull of what are in fact genuine moral considerations. Even though Pierre is mistaken about what the right thing to do is, and even though that mistaken verdict is the only explicitly moral belief he holds, nevertheless Pierre correctly identifies which features of his circumstances bear on the question of what the right thing to do is. He deliberates intelligently with his moral reasons, even if he ultimately draws the wrong conclusion. He has a good, if partial, moral understanding of his circumstances of choice.

The explanationist’s strategy has no traction in this version of the case. Pierre’s moral understanding cannot be accounted for in terms of his intermediate moral beliefs, since Pierre holds no intermediate moral beliefs at all. The problem of partial moral understanding, I take it, thus stands. In some cases, moral agents may have a good if partial moral understanding of their circumstances, even though there is no moral proposition about their situation such that they understand why that moral proposition is true.

I recognize that this very last example is somewhat artificial. I have made a number of assumptions, not only about which moral beliefs Pierre holds but even what moral concepts he possesses. My purpose in making these stipulations is not just to score a point against the explanatory picture. Rather, what the stipulations help us to do is to isolate an important positive lesson which can otherwise be hard to discern: that there is a kind of moral understanding involved simply in correctly taking morally relevant features of one’s circumstances to bear on what the right thing to do is.

If this form of moral understanding is not explanatory, what is it about? In the next and final section of the paper, I provide a brief account of the object of such moral understanding.

4. A Practical Answer to the Object Question

In the previous sections, I outlined two problems for the explanatory picture of moral understanding. The first kind of problem involves agents who have a sound explanatory understanding of why their actions are morally right or wrong, but fail to understand how those moral verdicts bear on what they should do. Cases of partial moral understanding, on the other hand, raise a problem with the opposite structure. They involve agents who manifest moral understanding while lacking explanatory understanding. As each of these problems illustrates, having moral understanding is not just a matter of understanding the explanation of some moral fact.

What else does the object of moral understanding consist in? Here is what I take to be a plausible answer: a central object of moral understanding is the reason relation between a proposition and an action. The reasons I have in mind are normative reasons. One useful gloss on normative reasons is that they are considerations that bear on a question.¹⁹ A reason to believe p, for instance, is a consideration that bears positively on the question of whether p is true. A reason to φ is a consideration that bears positively on the question of whether to φ. Drawing on this way of construing normative reasons, we can say that what it is to understand the reason relation between a consideration, on the one hand, and a course of action, on the other, just is to understand how that consideration bears on the question of what to do.²⁰ Thus, if some consideration counts in favour of φ-ing, then understanding how it bears on what to do requires treating it as a reason to φ; if a consideration renders a course of action ineligible, then understanding its normative import consists in ruling out that action as a live option because of it; and so on and so forth. Call the view that a central object of moral understanding is the reason relation between a proposition and an action the ‘Practical Answer’ to the Object Question.

The Practical Answer is a partial answer to the Object Question. I am not claiming that the only object of moral understanding is the reason relation. Similarly, I am not denying that understanding why certain moral facts obtain are genuine instances of moral understanding.

Instead, the Practical Answer can be thought of the conjunction of two positive claims. The first claim is that, in some instances of moral understanding, what is understood is the reason relation between some proposition and an action.²¹ The second claim is that these instances of moral understanding are especially central or important: they weigh heavily when assessing someone’s overall understanding in the moral domain.

In defense of the first claim, consider the two problem cases we introduced earlier. The Practical Answer has no difficulty accommodating cases of partial moral understanding. This is because the reason relation is not factive with respect to deontic verdicts. P may be a reason for S to do A even if doing A is not, all things considered, the morally right thing to do. This feature of the reason relation enables us to make much more fine-grained assessments of moral understanding than the explanatory account allowed. In Trip, Anna reaches the wrong conclusion about what the right thing to do is, but she nonetheless manifests a good understanding of her reasons: she correctly identifies the morally relevant features of her circumstances (that evading questions involves a degree of dishonesty, that simply telling the truth will make her family worry), and understands, reasonably well, how those considerations bear on what to do. In short, the Practical Answer allows for agents to manifest moral understanding even when they make the wrong decision about what to do.

The Practical Answer also has no difficulty explaining the failures of moral understanding exhibited in cases of bad reasoning. To understand how some consideration that is a reason against doing A bears on what to do, one must at a minimum take that consideration to bear negatively on the question of whether to do A. The wrongness of an action provides strong, perhaps always decisive, reason against that action. In the case I described, the hedge fund manager doesn’t treat the wrongness of hiking the price of the medication as counting decisively against doing so. They don’t understand how the wrongness of hiking the price of the medication bears on what to do, which is why they lack moral understanding.

The case of the hedge fund manager also lends some support to the second claim made by the Practical Answer, namely, that instances of moral understanding whose object is the reason relation are especially central or important. When we reflect on this case, it’s clear that no amount of explanatory understanding, no degree of depth and sophistication in this agent’s understanding of why some particular courses of actions are wrong, could compensate for their failure to appreciate how the wrongness of their action bears on what to do. Their moral understanding would remain just as deficient. This suggests that understanding how certain considerations bear on what to do is not only a genuine type of moral understanding, but also one that weighs heavily when assessing someone’s moral understanding in a particular situation.

More generally, the basic rationale for the Practical Answer is simply that it is a distinctive feature of morality that entities in the moral domain are sources of reasons. Properties like rightness and wrongness, but also persons, as well their rights and interests, provide strong reasons for action to all moral agents. This is not an extrinsic characteristic of the moral domain, but rather one of its defining features. We should thus expect an account of what it is to have understanding in the moral domain to accommodate that central feature of its subject matter. The answer to the Object Question that I have sketched is put forward in that spirit.

What about moral understanding that doesn’t pertain to one’s own action directly, but instead is about, say, a friend’s kindness, or the injustice of a law? It seems to me that in these cases too there are implications for how one should act. A friend’s kindness is a reason to emulate them in one’s own behavior, the injustice of a law is a reason to oppose it, and so on. Even when there is nothing one can do about an unjust law, say if the law belongs to the distant past, one may still have reason to oppose any law that is like it. Thus, even when one’s moral understanding doesn’t bear on one’s own action directly, such as when it concerns a virtuous character trait or the justice of institutions, deepening one’s moral understanding may still involve understanding how these facts about one’s environment bear on what to do, either hypothetically or in the actual world.

I have been arguing that one important object of moral understanding is the reason relation. But what is it to understand the reason for action relation? How are we to complete the ‘___’ in ‘S understands how R bears on what to do iff ___’?

That is an important question, but not one that it is the aim of this paper to answer. My aim in this paper has been to make progress on the question of what is understood in moral understanding. It is a further question what it is to understand whatever the object of moral understanding happens to be. Note that this further question was already a hard question on the explanatory picture too. After all, the nature of explanatory understanding, or what it takes to understand why p, is one of the central and most-debated questions in the epistemology of understanding. Providing a full account of normative understanding, or what it takes to understand how some consideration bears on what to do, is thus beyond the scope of this paper.

Nevertheless, here are what I take to be a few plausible features of normative understanding. Two minimal conditions are that, for S to understand how some consideration R bears on what to do, it must be true that R, and S has to be aware of R. In addition, S has to conceive of R as being having some importance, as being something that matters. To illustrate, suppose it’s Sunday morning and you’re deliberating about whether to stay in bed or go out jogging. Hearing the sounds of moving trucks outside, you then remember that this Sunday morning is the Sunday morning you’ve promised to help your friend move. Thoughts of snoozing in bed and jogging fade out. You’re now trying to decide what clothes would be best to wear for moving furniture, whether you can make it to your friend’s house on time, and so on. Given how remembering your promise directs your attention and shapes your reasoning about what to do, it’s clear that you conceive of that promise as being important. This brings us to a related condition on normative understanding. In order for S to understand how R bears on what to do, S must be disposed to use R correctly in deliberating about what to do. Note what this condition does not say: it does not say that you have to deliberate before acting in order to understand the reasons relevant to your action. Perhaps, in some cases, we make decisions for reasons without explicitly deliberating about the decision at hand. (Indeed, the case I have just given can plausibly be described along those lines: you’ve decided to keep your promise to help your friend move without deliberating about whether to keep that promise.) Even when we make decisions without deliberating, we can still be guided by our conception of what is important. So normative understanding doesn’t require deliberation. But it is importantly tied to deliberation: if you understand how some consideration bears on what to do, you are disposed to use that consideration well when you do deliberate.

Finally, one might wonder whether we have simply come full circle here: that what I have been calling normative understanding—understanding how some consideration bears on what to do—is just explanatory understanding in disguise. The worry, that is, is that when someone understands the reason relation between a proposition p and an action A, what they understand is why p is a reason to do A.

This objection overintellectualizes normative understanding. Someone can understand how some fact bears on what to do, appropriately conceiving of it as important and treating it as a reason in one’s deliberation, even if one doesn’t possess an explanation of why it provides such a reason.

In some cases, this is because there may not exist an explanation of why some fact is a reason. Some facts about reasons may in this sense be fundamental. For instance, suppose Konstantin wants to marry Kitty because he thinks his life will be happy and meaningful if he does, and miserable if he doesn’t. It is not implausible to think that there is no explanation for why the fact that Konstantin’s life will be happy and meaningful if he marries Kitty (call this fact ‘K’) is a reason to marry Kitty. If there does not exist an explanation for why p, no one can understand why p. So if there is no explanation for why K is a reason for Konstantin to marry Kitty, Konstantin cannot understand why K is such a reason. But, of course, even if there is no explanation for why K is a reason for Konstantin to marry Kitty, Konstantin can still understand how K bears on what to do.

Similar points apply even in cases where there does exist an explanation for why some fact is a reason. I take it that, in the example I gave above, you understand how the fact that you promised your friend to help him move bears on what to do. Plausibly, there exists an explanation for why that promise generates a reason. Perhaps the promise generates a reason because you cannot rationally will a world in which people wiggle out of their promises whenever it is convenient for them. Or perhaps the promise generates a reason because the rule of keeping one’s promises is optimific. Or perhaps the correct explanation is something altogether different. It would be nice to possess the correct explanation for why promises generate reasons, whatever that correct explanation is. But even in the absence of possessing the correct explanation for why the fact that you have promised someone to do A is a reason to do A, you can still understand how that promise bears on what to do. In short, normative understanding does not require explanatory understanding.

5. Conclusion

In this paper, I have aimed to do three things. I introduced a question that has been neglected in recent work on moral understanding: the Object Question. I argued that the most widely assumed answer to this question—the view that what is understood in moral understanding are explanations of moral facts—is incorrect, or at least incomplete. Finally, I put forward a different, albeit partial, answer to the Object Question, according to which one central object of moral understanding is the relation of normative support between a proposition and an action.