Replies to Driver, Johnson King and Markovits

I am extremely grateful for these thoughtful and challenging discussions. It is a great privilege to be able to engage with such sharp and insightful critics. In what follows I cannot address every point made, but will do my best to defend my view against the main objections and worries raised.

Let me start with Johnson King, who is very worried about my account of grasping Morality. It may help to begin with a bit of context. Why do I dip my toes into the murky waters of meta-ethics when my main project is clearly normative ethics and normative responsibility theory?¹ The answer is that I want to give an account of what sort of blame there can be in cases of moral ignorance. Obviously if there is moral ignorance, there must be moral knowledge. So we need to say something about what sort of thing moral knowledge is. My point in introducing the idea of grasping Morality (capital M denoting broad correctness) is to distinguish between two kinds of moral ignorance: moral ignorance that is deep—someone who is not even on the bus—and moral ignorance that is the mundane, everyday sort we encounter all the time. My basic picture is that if someone is not even on the bus, if they have been raised without any glimpse of the true morality, then we cannot blame them in the ordinary way (though we can still blame them in the detached way). However, if someone is in the same basic moral framework that the rest of us are in, we can blame them in the ordinary way. This is not because they are blameworthy for their moral ignorance, it is not that they really do know that their act is wrong, and it is not because we can trace their ignorance back to some previous culpable benighting act, but because there is a plausible sense in which they should have known better. That’s what seems crucial to me to the aptness of ordinary blame—we must be able to say to the wrongdoer, although you did not know, there is a plausible sense in which you should have known.

So what does it mean to say that someone doesn’t know but should have known better? Johnson King rightly urges that I should have said more about what exactly that means. She focusses on a sort of case that, in fact, I think is vanishingly unlikely, but I need to say a little more in order to clarify my account here. First, let me stress that it is true for most people that when they act in moral ignorance, they should have known better. And I don’t mean, people like me, educated university professors from the Anglo world—I mean most people in the world—people from all walks of life and all places. Most people share a basic ethical framework, and although ethical theorists may get very caught up in their differences, at some level (I submit) they know they are dealing with ethical differences within a shared framework.

As Johnson King herself says: “Speaking for myself, I have interacted with and learned from plenty of hardcore Kantians and Utilitarians, and I maintain personal relationships with people who harbor deeply-ingrained biases that I don’t think I can change their minds about.” That is exactly my point—we can maintain relationships with people with widely diverging ethical views, and with people who have blind spots and biases. They are not moral monsters. They are trying their best and their foundational views do not seem outrageous. Otherwise wonderful Aunt Bertha may be a bit homophobic. A dear friend is a hard core Kantian. My lovely neighbor voted for Trump. These are ordinary people making ordinary mistakes.

The thought is not that people making moral mistakes already know, in the way that someone might know that their spouse is cheating, but pretend to themselves that they do not know. It is rather that a better answer is available from the basic starting points that the agent accepts, and that if the agent had tried harder, they would have been able to reach that answer. Furthermore, there is no particular route to knowledge that must be possible in order for my claim that the agent should have known better to apply. My thought is not that I should be able to change their mind—or indeed that anyone should. Rather, the thought is that at some level they could come to the correct view from their current framework. It may be that there are psychological barriers (laziness and bitterness are common ones) to them doing so, but the basic framework that they accept points to a more accurate moral view than they actually have on some topic. And by contrast, someone like JoJo, who is outside of our moral community now, could be brought in by extensive education. He couldn’t get there on his own (it’s not that if he had tried harder he would have come to a better answer), but he could get there with help. Finally, whether or not someone can come to the correct view is not a matter of the content of their view—so it is not that there is a distinction between sexist and racist and homophobic belief sets, such that some are more outside Morality than others. I think most people who hold sexist or racist or homophobic views should know better, and their beliefs are a result of laziness or blind spots or self-deception. They could and should have tried harder. The problem areas are not well integrated with the agent’s basic outlook, and the agent is failing to see that.

The view I defend is a threshold view. At some point, an agent counts as having enough of a grasp on Morality to be in the moral community, and subject to ordinary blame. So what is the threshold? Here, I don’t say much, but I think I say about as much as I can say. We can see the extreme cases of people who are not in the moral community, usually fictional characters: JoJo, and Bill (who is raised in a completely isolated community). And there are the easy cases of those who are in the moral community (most people). But there are borderline cases. I take Huck Finn to be a borderline case, as I say in chapter 2. There are other potentially borderline cases, cases where an individual or a whole society is in a state of transition from one morality to another. I don’t have a firm view about such cases. But I think that is the right attitude to have: we don’t have to shoe-horn a complicated situation into a neat theory.

Johnson King’s concern is not so much with the borderline cases, as with people who do grasp Morality, but have gaps in their moral knowledge. As I say, I think that is very common. People may miss things, have blind spots, fail to make inferences, suffer biases, and so on. Yet, on my view, they count as grasping Morality, and so we can blame them in the ordinary way. Johnson King objects that someone who acts on local ignorance believes that they are acting rightly, so in a sense, they are acting subjectively rightly.

First, I should note that on my view, things may be so complex that two people could both count as acting subjectively rightly even though they come to different conclusions. This is where the notion of trying is important: you count as acting subjectively rightly when you are trying to do well by Morality, and of course sincere trying may not result in a true belief. But so long as you are trying within the right framework, and it really is sincere, that can render the action subjectively right. So some ways of being morally mistaken do count as subjectively right.

Another way to be mistaken and not blameworthy, is to have a simple excuse. As I acknowledge, it is very unlikely that the ignorance of ordinary agent is non-culpable, and that there is no sense in which they should have known better. It is very difficult to think of a good example of a simple excuse, and this is grist to my mill (as I will explain later). So here, instead of an example, if the form of a case of unavoidable non-culpable moral ignorance. The agent generally has the right sort of moral outlook, but in some isolated case, holds a position that is not just incorrect, but unreasonable given their background beliefs. However, it is not so unreasonable that they could be expected, as a being with normal cognitive capacities, to see that it is unreasonable. There are lots of non-moral facts that we get wrong in this way. I have the right scientific world view in general, but I am sure I have many false and unreasonable beliefs about how things work. Some of these are culpable (I am too lazy to look up how yeast works), and some may be unavoidable (given how I was taught and what everyone else believes, I may believe that lactic acid build up causes delayed onset muscle pain. But actually that is not true²). It is harder to think of isolated but unavoidable mistakes I might be making about Morality. Every example invites the suspicion that the agent should have thought about a little harder.

So, (as Johnson King suggests I should) I think that most cases are cases where the agent does not have a simple excuse, and where they should have known better—they should have tried harder—and so we can blame them. This is where Johnson King sees a problem. As Johnson King emphasizes, the agents who suffer local moral ignorance think they are acting rightly when they act on their ignorance. And my account of subjective obligation and ordinary blame depends on the thought that you are acting wrongly by your own lights, and that you could come to see that your act is wrong. As Johnson King puts it, “we are dealing with precisely the cases in which agents with gappy grasps cannot be “reminded” of the wrongness of their action because they do not know (and may never have known) that it is wrong. We cannot remind someone of something they do not know (and may never have known). So, since ordinary blame is just a reminder—as Mason holds that it is, per claim (e)—it turns out that these agents cannot be subject to ordinary blame after all.”

The missing link here is my claim that to be acting wrongly ‘by your own lights’ is not necessarily seeing clearly that you are acting wrongly. The light is not a searchlight, as the contrast with the ‘searchlight view’ (so called by its detractors, not its supporters) indicates. Rather, the lights in the metaphor can be very dim. Indeed, acting wrongly by your own lights is compatible with believing that you are acting rightly. The crucial claim is that the agent’s own general moral outlook points (not necessarily obviously) to the right answer. So, for example, someone who has racist, sexist or homophobic blind spots is usually blameworthy in the ordinary way because the rest of their moral outlook should (could without massive intervention) lead them to the conclusion that excluding some category of people from moral concern is not compatible with Morality. It may not be easy to convince them of this, and it may be that they resist being led down that path. But still, there is a sense in which they should know better.

Why might we think that these gaps are not the sort of case where it would be fair to say that the agent should know better? My exposition of grasping Morality is at fault for slipping into the claim that in some sense, an agent does know better. Johnson King focusses on the idea that an agent is reminded of what of what she already knows, and I do put it that way at points (unwisely, I now see!). A better way to put my point is that the agent could have come to the right view, “if they had thought about it a little longer, if they had thought more about their dispositions and alertness to red flags, if they had tried harder, in other words.” (p. 90). So on my view, the sorts of gap that those who grasp Morality have are not deep gaps, rather, they are shallow gaps, gaps that could be filled by trying harder.

The background thought here, as I acknowledge, is that Morality is not a series of logically unconnected dictums, there is a sort of pattern and coherence, and so although there may not be a strict implication from other beliefs an agent has, there are pressures of coherence towards seeing that (for example) people with the same morally relevant properties have the same moral status and that things like skin color are not morally relevant properties. This is why it is hard to think of examples of isolated moral ignorance (gaps) that are non-culpable. Imagine, for example, that someone’s attitudes to animal rights have been formed in a cultural context of agriculture and poverty. It might be suggested that this warps one’s view of animal moral status in a way that is unavoidable, and in that case, the ignorance is non-culpable. But given that the rest of the person’s moral outlook is on track, it is not plausible that the ignorance is unavoidable. There are not too obscure routes from the other tenets of morality to the view that animals have some moral status. Morality is fairly coherent.

I think this last point is where there is a disagreement. Johnson King must be thinking of these knowledge gaps as deep gaps, gaps that cannot be filled by appeal to coherence with the agent’s other beliefs. And this of course is a very complex disagreement about the nature of Morality, which is beyond the scope of the book. But I will nonetheless attempt to give one more persuasive example, using the inverse case, that Johnson King presents at the end: she imagines someone who is mostly gap, and counts as outside Morality, but has some islands of moral knowledge. If this person tortures kittens knowing that it is wrong, they act wrongly by their own lights and (Johnson King argues) should be blameworthy. But on my view, as Johnson King points out, they are not in the realm of ordinary blame. My response to this is that it is just very hard to think of someone with so little moral knowledge as having real knowledge on the islands. What are we to imagine here? Remember that on my view, lacking a grasp of Morality is unusual, and extreme. So perhaps we are imagining someone who believes that it is ok to torture all kinds of other beings, but not kittens. In that case, it seems to me, they have not really grasped the wrongness of torturing kittens. Similarly, if someone really does grasp Morality, it is hard to see that they could have really deep gaps of genuine ignorance. But if they did, of course that could be non-culpable ignorance, and they would have an excuse, as I say.

Some of these same points are relevant to Julia Markovits’s comments. Markovits objects to my view that when we blame someone in the ordinary way, the point is to communicate with them, and expect some kind of uptake and response. On my view, which I admit is overstated in the line Markovits quotes, a communication can be successful because it reminds the agent of something she already knows. As Markovits points out, that is way too high a bar for felicitous communication. More carefully stated, my view is that there must be some route from the agent’s general moral outlook to seeing the wrongness of the act we blame her for. If an agent is deeply ignorant—they don’t share any of our basic moral principles—then we could not expect them to understand why we are blaming them.

Markovits argues that even people who are fully morally ignorant (such as JoJo) or very ignorant (such as Huck Finn, who I see as a borderline case) may be susceptible to blame despite having no or little footing in Morality. As Markovits puts it, “Blame may emphatically draw its target’s attention to moral claims others have on him or to their moral standing. It can effectively persuade a misguided moral agent to reconsider his commitments.”. I agree with this. But there are different sorts of case here. Take first Huck, who I agree may be construed as having a grasp of Morality, or at least, being close to the threshold. When we think of Huck like that, it makes sense to blame him in the ordinary way. I think that there are many cases where there is a complex mix of bad beliefs and good motivations, or good beliefs and bad motivations, and such cases are not easy to theorize. But they are not easy to respond to in real life, and our theories should reflect the complexity rather than try to iron it out. Huck is a complex case. So are the cases of Kleinbart and the akratic adulterer that I discuss later on. These cases are described in such a way that there are elements of both self-aware and non-self-aware morally valenced behavior, so there is pressure to both ordinary and detached reactions. In my view—the view I argue for in the book—the best way to deal with this is by acknowledging that these cases are mixed, rather than arguing that we should focus on only one part of the story.

Let me return to JoJo. I agree with Markovits that we may hope that JoJo could be convinced to change his views, and that we may act as if we are blaming him in the ordinary way. But I do not think that this is ordinary blame. Rather, I think insofar as it is blame it is proleptic blame—not in William’s sense of giving the miscreant a reason (as Markovits helpfully clarifies)—but in the sense of treating someone as if they were in the moral community in order to draw them in. And it may not be blame at all, but straight up moral education.

Consider how we might talk to JoJo. Imagine that (after a hefty donation from his father) JoJo arrives on campus to study for his degree. JoJo, as he is usually imagined (and as I imagine him in the book) is fully indoctrinated with his father’s world view, so is fully outside our moral community, but has all the usual cognitive capacities. There is no point in saying things like, ‘hey, JoJo, I can’t believe you cheated on your exams! You have let the whole system down!’. Or even things like, ‘I’m shocked that you are so disrespectful to a woman professor.’. But, because we know that JoJo has all his cognitive capacities, and because we have faith that Morality is pretty compelling, it seems worth trying to convince JoJo to reconsider his commitments. We may do this by showing him our dismay at his actions, by treating him as if he will respond to our resentment. That would be proleptic blame. But, it seems to me, we may be better off just trying to explain to him how some of the basic principle of Morality work, and show Morality in its most compelling light. That would be straight up moral education.

One final point here. In cases like JoJo, we usually don’t know what the person’s grasp on Morality is. It takes some exploration before we could firmly conclude that they do not grasp Morality. Until that time, the charitable way to proceed is as if they do, and so in practical terms we should default to ordinary blame rather than detached blame. It may misfire of course.

Markovits’s larger worry here is that my account of the conditions for ordinary blameworthiness leaves too many people out. Markovits agrees that Huck and JoJo are disputed cases in philosophy, but she thinks that are much less controversial cases that my view would exclude from the realm of ordinary blameworthiness. She mentions Paul Jennings Hill, who murdered an abortion provider and his bodyguard; those who might perpetrate injustices on utilitarian grounds; those who give less to charitable causes than they should do, while really convinced (even deep down) that they’re giving enough, as well as politicians with false views.

First, those who commit extreme moral crimes are always puzzling, and it is always tempting to think that such people are mentally ill to the point of not being responsible at all. If mental illness is ruled out, there are different ways to fill in the story, and I think that different ways we fill in the story make a difference to what my view will say about the cases. Was Hill really deep down convinced, or was he self-deceived?³ In cases where someone is moved to violence in the name of their beliefs, it often seems as though they have been carried away with passions that feed their self deception. When I watched the footage of fired up rioters at the Capitol on January 6th 2021, I found it hard to believe that they really believed the things they claimed to believe. They seemed ‘carried away’. Surely that can happen in an isolated individual case too (of course, such people are not isolated, they have the internet as their support mob). People allow themselves to be caried away by fervour, and they lose track of reality. On my view, this is self-deception, not deep moral ignorance. So Hill would be subject to ordinary blame on that story.

If we think of Hill as fully convinced of the necessity of killing abortion providers, but otherwise in the realm of Morality, then this is the sort of case that Johnson King worries about—there is a gap in his moral knowledge, but one that is so deep that it cannot be filled by appeal to the agent’s other background knowledge. My answer then is as above: I do not see that this is a plausible description. Even if someone’s moral view exaggerates the moral status of fetuses, how can someone who grasps Morality non-self-deceptively fail to see that adult human persons should not be lawlessly killed by other citizens to protect fetuses?

This segues into the case of someone who makes the mistake of thinking that a utilitarian calculation licenses injustice. In general, I do not think that disputes between philosophical moral theorists count as disagreements about Morality: rather they are disagreements within Morality.⁴ Most utilitarians bend over backwards to avoid the conclusion that innocent people can be killed for the greater good. And those who may, for shock value or whatever, suggest appalling conclusions, would not act on such proclamations. So the general issue here is not so much people who make that mistake in moral theory, but people like Hill. And as I say, I think we can fill out that story in different ways, but none of them render my account or ordinary blameworthiness too exclusive.

Markovits says various interesting and compelling things in favor of the moral concern account, and I can’t address all of them. I do see the force of Markovits’s worry that my account of detached praise and blame does not do full justice to the cases I expect it to cover. Markovits objects to my claim that detached praise and blame is not really concerned with the target’s motivation, but primarily with what they have done or how they are. Markovits, rightly I think, points out that when we praise or blame someone, we are interested in why they did what they did (and this of course counts in favor of Markovits’s moral concern approach to praise- and blameworthiness).

I concede that even detached praise and blame focus on motivations, for exactly the sorts of reason that Markovits gives. But I remain unconvinced that we can interpret cases of non-morally motivated action (concern that is not self-consciously moral) as praiseworthy in the same way that action that is motivated both my concern and moral considerations. I don’t have much more to say here, but let me use Markovits’s own example, the nut-allergic friend. Markovits argues that concern for a friend’s welfare is enough to mark them praiseworthy (minimally so of course) for avoiding giving them nuts. And I agree, at the everyday level of motivations. But we all know that it would be wrong to give them nuts because it would hurt them. And if we were to ask the agent—do you know it is wrong to hurt people?—we would of course expect them to assent. In the normal everyday case we don’t need to ask that. But when we think of a case where the agent replies, ‘no, I did not know that, or ‘no, I don’t really are about that’, we are much less comfortable with ordinary praise.

Julia Driver takes issue with my account of blame. On my story, there are three different kinds of blameworthiness, and correspondingly, three different kinds of blame. The central distinction is between ordinary blameworthiness/blame and detached blameworthiness/blame. On my view, when someone who grasps Morality—someone who is in our moral community and therefore ‘should know better’ in a reasonable sense—acts wrongly, we should blame them with ‘ordinary blame’, where ordinary blame is communicative. Driver sees various problems with the idea of communicative blame. She argues that blame is not always communicative (it is sometimes private) and it does not always issue demands.

Let me first address the worry that there can be private blame. Driver’s thought is an intuitively attractive one, that much blame is private. And, she argues, we cannot argue that private blame is simply silent communicative blame, because sometimes there is no wish to communicate, in fact, it might be that blame is “expressive of as desire to have no interaction at all with the wrongdoer”. Of course I do make room for non-communicative attitudes (such as contempt and disdain) in my account of detached blame, but Driver is talking here about cases where the wrongdoer is clearly in the moral community and acting wrongly in the ordinary way, so I cannot respond by appealing to detached blame.

I agree that it is very common, when one has been wronged, not to want any contact with wrongdoer. Lots of very personal wrongs, the kind we could call ‘betrayals’, sting in a way that makes communication difficult, and a very natural response is to withdraw and banish, or at least, avoid, the wrongdoer. But should we think this is not a communication? As children in the school playground know all too well, exclusion is a powerful communication. And (unlike in most cases deployed by children), exclusion and banishment are often justified in the face of betrayals. The message being sent is, ‘you have hurt me too much for me to talk to you’. I think that that is consistent with another, implied message, which is, ‘you should be grovelling, but I am so angry and hurt I don’t even want you too’. There is a demanded response, but it is demanded only in principle, in practice, it is waived. In other words, there is a way to model private blame on communicative blame.

Sometime of course, and often in cases of betrayal, we do retreat into contempt and disdain, which are not forms of ordinary blame, precisely because they do not demand a response. I think that what is happening here is that, as Strawson suggests we sometimes may, we are retreating into a more objective attitude because we are under too much strain to keep engaging in ordinary communicative blame (1962). Even when someone is eligible for ordinary blame, we may decide we can’t face it. And I think that is a recognizable and distinct way to approach a wrongdoer. Sometimes, ordinary blame demands too much of the blamer.

This account of blame includes the thought that blame makes demands. It is because blame makes demands that ordinary blame is demanding for the blamer too. So it is not surprising that Driver also worries that there can be a sort of blame that does not make demands. And, as I say above, I agree. I think that we can sometimes retreat into detached blame in order to avoid the burden of making demands. But Driver has another sort of case in mind too. As she has argued in more detail elsewhere, it is possible to be morally at fault, and blameworthy, without having acted wrongly (1992). Driver points out that the sort of blame that is apt when someone has acted suberogatorily cannot be understood on the demand model—‘Morality demanded that you do P, you did Q, and now my blame is a demand that you make amends.’. (Notice there are two different positions for demands here).

This is an interesting case, and I think that Driver’s own suggestion about how to respond (which, as she acknowledges, I gesture to but do not expand on) is on the right track. The idea is that when the problematic act is just suberogatory rather than a failure to meet a moral demand, moral criticism rather than blame is appropriate. Driver worries that this will not be able to accommodate the anger that may be warranted in such cases (as when someone’s sister refuses to donate their bone marrow, for example). But couldn’t we say that the anger is not moral anger, and so not part of the moral reaction? If someone needs a bone marrow transplant, and their sister stands on their rights and refuses to be the donor, the sort of anger we might expect is not moral, but relationship based. So the appropriate reaction includes both moral criticism, and anger about the lack of loving response. Unfortunately, many of us do have a tendency to moralize, to treat failures in others love for us as moral failures. But love—like sexual attraction—is not morally required, and the anger and disappointment we feel when we don’t get what we want in these areas should not be moralized.

Driver is also unpersuaded by my account of taking responsibility (although I appreciate the suggestion of a possible extension of the view to cases like those discussed by Sarah Paul). In brief, I think that one can take responsibility for inadvertent action, and that one does it in order to show respect and commitment in personal relationships where one has inadvertently harmed another. Driver argues that this is not what actually happens. In fact, what usually happens is that as soon as an apology for inadvertent harm is offered, the other party will, and should, deflect the apology. Here is how Driver puts it in the case of the friend who has lost a previous necklace, and apologizes:. “… it also would seem to me that her friend should respond as follows: “Oh, Perdita—no need to apologize, I know what a bad day you had yesterday! Let’s just go out and you can treat me to a pizza.”. My question here is, what is the role of the pizza? Driver thinks that the friend should point out that apology is not appropriate. But if that is the case, she shouldn’t ask for pizza either! Of course, Driver is likely to respond that the pizza could equally be paid for jointly, that part of the description is not important. But, I think it is important. It does seem apt that Perdita should buy the pizza. Just as she should apologize. And, crucially, accept that the friend’s resentment (such as it is, it may not be much) is aptly directed at her. The point is that there usually is some resentment, even when the harm done is clearly inadvertent, and it behooves us to take it on when it is directed at us by the people we love. That may mean simply buying a pizza.