▶︎ 0:01 Welcome to lecture four on moral discourse. So far, I've talked about the ways in which I think metaphysics, especially in the years since 1950, has become dysfunctional, and that it's become dysfunctional in part by not paying attention to the different functions that language serves, and instead treating all language on the model of the kind of observational language that we learned first.
▶︎ 0:28 I've also talked about how we can do better by thinking of metaphysics as a kind of conceptual engineering, including looking backwards, reverse engineering to understand the functions of various areas of discourse, and constructive or forward-looking engineering. And then I tried to talk about how we can understand linguistic functions and how this makes a difference to debates about properties and numbers.
▶︎ 0:50 What I'm going to talk about today is another huge topic in the history of philosophy and the history of metaphysics, and that is morality. Or, as I should say here, because the whole theme is to start by taking a step back and asking how the discourse works, I'll start by analyzing moral discourse. Work in the metaphysics of morality is in the subfield that's called metaethics.
▶︎ 1:17 Metaethics doesn't ask questions like, what should we do? How should we treat others? What are the basic human rights? Instead, they ask metaphysical questions, and sometimes questions about knowledge, epistemological questions about morality. Questions like, are there, or are there really, moral facts, moral properties, moral obligations, and moral requirements? If so, how could we know them, for example? And what are they? What kind of things would they be, the metaphysicians want to say.
▶︎ 1:52 How could they be placed in the natural world? And if they are parts of the natural world, wouldn't they have to be really strange kinds of things? Because the whole idea of a moral obligation is it's something that ought to motivate you to act in certain ways. But how could a mere thing in the world prompt you to action, is one of the questions that's often raised. And how can they be grounded in other physical properties of the world to make something morally right or wrong, required or optional, whatever?
▶︎ 2:24 These kinds of problems and concerns about whether there really are moral facts, moral properties, and so on, they're actually really important because some people have raised the concerns that we just can't make sense of these things, that they'd be strange kinds of things that we couldn't place in the natural world, or that we couldn't have evolved any kind of faculty to come to know them. And if you start to take that kind of view seriously, this can lead to a moral skepticism or even a nihilism that says there's really no moral obligations or nothing that we have to do. And that would be a big change in our worldview, at the least, and maybe also lead to a large range of behaviors that people just don't bother restraining themselves from anymore if they think there aren't really any moral obligations against them. So these are metaphysical questions, but they're also really important questions that are relevant to how we live together.
▶︎ 3:22 As in the other cases, the view I want to take to approach these problems is one that begins by starting with a step back, by asking questions about moral discourse, what functions it serves in human life, how it can get introduced to human language, what sort of rules govern its use. And again, I'll try to show you that if we take things in this direction, then we can come to see that some of the problems in the history of metaethics, the metaphysical end of ethics, arise from mistaken presuppositions, and others can be easily answered.
▶︎ 4:03 At this point in the lecture series, there's three points for me doing this. One is it shows another important application of this functional approach to metaphysics and what we can get out of it, and one that works rather differently than the talk I gave yesterday about nominalizations and how they work in property talk and number talk. I'm also going to use this as a way to introduce a kind of approach to functional analysis of language that takes us step by step through different grammatical developments and shows how different functions can get layered on to sophisticated forms of talk.
▶︎ 4:43 Finally, it'll give a kind of answer to how I think about practical questions. If I'm right, then a lot of the questions we think of as metaphysical questions, questions about the world, are covertly practical questions about how we ought to think and talk or sometimes what we ought to do. How can these practical questions be resolved, and can they be addressed in a way that doesn't already presuppose the deep worldly view of metaphysics that I've been arguing against? So those are three points to this line of discussion, given where we are and where we're going.
▶︎ 5:19 Let's get into metaethics then. I mentioned in the very first lecture that recent work in metaphysics has often been plagued by a kind of blindness to functional questions and a blind assumption that all discourse functions in the same way, whether to posit explanatory features of the world as in the Neo-Quinean view, or to track and describe features of the environment. Now, metaethics is an exception to this. This is one subfield where functional questions have actually been discussed a lot and have played a large role in work in this area in the last, let's say, 100 years.
▶︎ 6:03 And there have been then many past views about the function, often it's put in the singular, of moral discourse. So back in the early 20th century, there were what were called prescriptivist views. They were defended by people like Carnap, who I mentioned in an earlier lecture, also by Hare, by Stevenson. Roughly, they thought the function of moral discourse was like the function of a command to prescribe, to tell people what to do, not to describe some facts of the world.
▶︎ 6:38 There's also been a lot of very sophisticated expressivist views of moral discourse, views that say, look, the function of moral discourse, again, isn't to describe the world and how it is or describe the moral facts and properties. But most basically, it's a function of expressing our attitudes towards different types of action, for example. Simon Blackburn has been really a key sophisticated defender of this approach. Alan Gibbard has his own version of it.
▶︎ 7:07 And then there have been the generally serious moral realists who will take a descriptive view of the functions of moral discourse. But at least they take time to defend this view because they can see the alternatives. And they will hold that really the function of moral discourse is to describe moral facts and properties and to state truths about them and communicate truths about them. So we get at least three kinds of answers to the question of the function, again, assumed singular, of moral discourse in past prescriptivist, expressivist, and descriptivist work.
▶︎ 7:43 Now, that in itself has led to some kinds of skepticism. A lot of people will look at that and say, "Look, how can we possibly tell which of these functional stories is correct?" Everybody's got their own opinion. Maybe they're just using them to support their own metaphysical views, or getting rid of metaphysics if you're a prescriptivist or expressivist. How could we possibly know which is the real function of moral discourse?
▶︎ 8:09 What I want to say in response to that is we can actually make progress on this question by using the functional approach that I outlined in the prior lecture and drawing on work in linguistics to understand the functions of moral discourse and, notice the plural, the ways in which these functions get layered up through grammatically different forms of moral language.
▶︎ 8:35 Think about this to start with. Broadly, moral language can take many different grammatical forms. One of the most basic forms uses the mood system. Imagine you're parenting now, and you want to just say, "Don't hit your brother." So that's an imperative, but it's an imperative that's a way of trying to regulate the behavior of the child who's hitting. But we can also make moral statements with modal terms. We can say, "You mustn't hit your brother." Probably best if you don't put it in those forms until they're four or so and have got modal discourse. But you can say that, too. You can say, "You mustn't hit your brother."
▶︎ 9:18 We can also introduce nominalizations and talk about moral facts, moral properties, moral requirements, moral obligations. So then we can say something like, "It is a moral requirement to refrain from physical abuse." So we can communicate moral ideas in various different grammatical forms. And one thing I'm going to try to show is that discourse in each of these different forms serves different total sets of functions.
▶︎ 9:48 And once we see that, we can understand also how to preserve the grain of truth behind prior functional views, behind the prescriptivist view, the expressivist view, and even the descriptivist view as part of a total view about the many different functions that moral discourse of many different forms serves. Then we can reconcile those views and get a much more complex and sophisticated view of moral discourse, which still, I'll try to show at the end, sheds new light on the old metaphysical problems and can help us either answer them easily or show why they were based on mistakes. So what we'll get at the end then is a fine-grained, multilayered assessment of the many functions of moral discourse of various forms.
▶︎ 10:36 Now, before we get to morality proper, I'm going to step back. The linguists that I've been reading, they don't really pay a lot of attention to the distinction between properly moral discourse and what you might call, or philosophers would call, other forms of normative discourse. In all sorts of areas, we talk about what you should or shouldn't do, what you must or mustn't do, what to do. Not just morally, but also what you must do in order to remain a student in good standing at this university. Or what you must do to be awarded tenure at this university, or what bus riders are required to do and not do.
▶︎ 11:20 We have all sorts of regulations. Call this normative discourse, generally. All of this aims, in some way, to regulate our behavior for some particular purpose or other. You might think moral discourse aims to regulate our behavior just more generally. At any rate, I'm going to start with understanding how normative discourse functions, and then we'll ask what else we get that's particular, that's distinctive in moral discourse.
▶︎ 11:48 The typical metaphysical questions that I mentioned at the beginning, questions like, are there moral facts, properties, or requirements? What sorts of things are they? How could we know them? Notice that these terms that are used in presenting the metaphysical questions, moral facts, moral properties, moral obligations, moral requirements, these are all nominalizations. They're all grammatical metaphors.
▶︎ 12:15 What we know already from the work on linguistic functions is that these are introduced very late in the developmental story and to serve certain particular kinds of additional functions. So it's very interesting that these metaphysical questions begin at what you might say is the developmental end of the story. We'll see why, I think, as we go on.
▶︎ 12:38 But in order to shed some light, I think it will help to take us step by step through different stages of the development of the discourse. This doesn't have to exactly track what each and every child does, but it is a conceptual priority that as one form of discourse builds on the other and can be introduced via the other. If we go step by step, I think we can better see how the functions get layered up. So let's take it back, and we'll start with just normative discourse generally, not worrying about what's specifically moral, though that will be included.
▶︎ 13:14 As I suggested a minute ago, one of the original forms of this discourse is discourse that's using the imperative mood, that requires the capacity to use a mood system to distinguish imperatives from declaratives and interrogatives and so on. The mood system, as I mentioned in yesterday's lecture, functions to enable speakers to distinguish imperatives from interrogatives and declaratives. Kids tend to learn about it at age 18, 19 months.
▶︎ 13:50 And with that imperative system, you can do a lot to regulate behavior. So you can say, "Don't hit your brother." Or in a closed context, like in a philosophy classroom, you can say to your students, "Read Kant's Groundwork for Monday." Uttering imperatives to tell them what to do or not do. That can go some distance towards helping us to regulate behavior.
▶︎ 14:23 The modal system enters a little bit later, around ages two to four. So the next question is, what does that enable us to do that we couldn't do with just the mood system? We can introduce deontic modal language to say things like, "You mustn't hit your brother," for example, or, "Students must read Kant's Groundwork." What can we do with a modal system that we can't do with just commands in a mood system?
▶︎ 14:50 Well, one thing we can do is help regulate behavior of other people without overtly taking on the role of the commander. I can be more polite, not barking orders at my students, but tell them what all students of Philosophy 101, what they must do. We can give generalized regulations and make it clear that these apply to anyone who's enrolled in the class, not just to whoever hears me and not to anyone who doesn't hear me. Presents them in a kind of impersonal and polite way.
▶︎ 15:21 Another neat thing that the modal terms do is it enables us to modulate the degree of force in certain ways. So I could say, for example, to a child, you can say, "You must clean your room," or more weakly, "You should clean your room." More weakly still, "You might clean your room." So we can modulate force in a way you can't with imperatives. You could either say, "Clean your room," or you can say, "Don't clean your room," or you can say nothing. But you can't subtly adjust the force in this way.
▶︎ 15:54 A really important feature of modal discourse that you don't get with imperatives is the ability to give permissions explicitly. So you can say, "Read Kant's Groundwork," or, "You must read Kant's Groundwork." But you can't give a permission like, students may retake one exam during the semester or may skip one reading question during the semester without having a modal verb like may. And that's really important to make explicit what's permitted as well as what's required.
▶︎ 16:27 Another really important thing that the modal system brings in is it enables us to regulate behavior with utterances that are in the declarative mood rather than the imperative mood. Now, why is that important? Well, it's important because we can reason in certain ways from statements in the declarative or indicative mood that we can't from imperatives. So, for example, I can say, "Read Kant's Groundwork," but I can't say, "If read Kant's Groundwork, then blah, blah, blah." But I can say, "Students must read Kant's Groundwork," and the students can say, "If we must read Kant's Groundwork, then this is going to be a hard course." Or, "If you mustn't hit your brother, then you mustn't hit your sister either," and reason with them in ways that you can't with imperatives.
▶︎ 17:21 You can't put an imperative in the start of an if/then statement. It doesn't make grammatical sense. Now, why is that important? That's important because then we can reason from one thing we're told to do to conclude other things that we must or mustn't do, that we may or can do, and we can get more generalized learning about what is permitted or required by reasoning from what's explicitly stated. That's, of course, not the only ways or the only set of ways in which we can communicate morally.
▶︎ 17:55 We can do imperatives, we can use modal verbs, but we can also introduce new nominalizations. And as I said in the prior lecture, this ability to fully understand and process nominalizations really only is fully achieved around ages like 8 to 12 for children, so it's very late. But what we can do is to move from talk of what one must do to talk of what is required, to talk of what's a requirement, or from what one may do to what is permitted, and to what permissions there are, introducing new noun terms. And that enables us then to speak of requirements and obligations, permissions, and so on.
▶︎ 18:43 Why is that important? Well, look, as in other cases, once we have these nominalizations, we can qualify and quantify them. We can talk about, say, the requirements for this course are more onerous than the requirements for the course I took last term. Or we can count them. We can say there are five requirements for this course. So we can do these kinds of things with them now. Once we've introduced the grammatical metaphor, that enables us to say more about the requirements and the permissions, to compare them, to enumerate them, to evaluate them, and so on.
▶︎ 19:19 So these kinds of nominalizations that get introduced very late, they're a very sophisticated form of normative discourse, generally, enable us to do new things. We can quantify requirements. We can qualify them, say this is an onerous requirement, or this is a stupid requirement, or whatever, and compare them. We can insist on justifications for these requirements. Again, there's another nominalization, a justification for the requirement. We can generalize and theorize now about requirements and reason about them, and about moral facts and properties in ways that are, again, central to bureaucracies.
▶︎ 19:55 You might want to know what are the requirements for becoming a homeowner in this development, for example, or for acquiring a college degree. And you can theorize about them. So think about a lot of the questions and claims in traditional moral theories. So you might make a generalized claim that's not just barking an imperative at somebody. But instead, if you're doing moral theory, you might come up with the idea that the severity of punishments should be proportional to the wrongness of actions, for example. Or you might ask generalized questions about, in virtue of what are there moral obligations at all?
▶︎ 20:39 Or you might come up with a generalized moral theory like Mill's statement of utilitarianism was, actions are right in proportion as they tend to promote happiness. Notice all the nominalizations in that sentence. You can't get a kind of generalization like that without having these nominalizations at your disposal. Or if you want to say moral requirements are independent of our desires. Again, you've got to have nominalizations there.
▶︎ 21:07 So these nominalizations of normative ways of speaking, including moral ones, they enter language relatively late. They are conceptually very derivative, but they're very useful. They are what's centrally useful to moral reasoning and moral theorizing. And that's why we see such a focus on them in work in philosophy, on ethics, and on metaethics. So it's no surprise that they focus there, even though these are sort of not the original forms of regulative discourse at all.
▶︎ 21:41 Just to summarize, we can get a layered understanding of these functions. You might say that it all originates from some shared human needs, to take a needs-based approach like Matthieu Queloz suggests. We are practical creatures, and we are social creatures. We have needs to decide what to do, not just individually, but collectively. What shall we do? How should we act? We have needs to regulate behavior so that we can better live together in society.
▶︎ 22:13 We can start to manage these needs with a mood system by giving imperatives and commands. In the mood system, we can regulate behavior with imperatives. We can also express attitudes, as Blackburn said. We can say, boo to hitting your brother, and express a negative attitude towards that.
▶︎ 22:37 But then as we get more linguistically sophisticated, we can do new things that help us regulate behavior better. We can introduce deontic modal verbs that enable us not just to command, but to explicitly permit various kinds of behavior. We can do so with varying degrees of stringency, whether you should or must do something, and show varying types of attitudes towards the action. But we can also do it in a declarative mood, so we can reason with and from it. So we can reason from, if you mustn't hit your brother, then you mustn't hit your sister either, for example.
▶︎ 23:11 And then we can get even more sophisticated and add nominalizations out of these modals, and they enable us to speak of moral properties, moral facts, moral obligations, moral requirements in thingly terms. And that enables us to generalize, demand, and give justifications for regulations, compare requirements, develop bureaucracies, laws, and moral theorizing.
▶︎ 23:38 So if you think a little bit back to the history of philosophy, you may know that Thomas Hobbes spoke of the commands of the sovereign and their importance in establishing a civil society and a government. One thing we can see now is why there's a whole lot that commands of a sovereign in uttering imperatives, ordering people around, can't do, that you can only do when you get more sophisticated forms of language, like deontic modal terms and even nominalizations. These are what's required to move from what you might call the rule of man, where you have one individual barking orders to the rule of law, where we have impersonal general systems of requirements and permissions that apply to everybody that we can reason with and from. So that's really important.
▶︎ 24:33 What I've said so far covers normative language, whether we think of it in terms of giving the rules of behavior in my classroom or for students at this college or in the laws of this state, or whether we think of it in moral terms. So you might want to ask at this stage, that was interesting. This governs all kinds of local forms of behavior or rules of games. Those are local norms, also club members, employees. But what is moral discourse? Moral discourse feels different than that. It feels more important and more universal than these kinds of local normative regulations.
▶︎ 25:10 Can anything more be said about the functions of properly moral discourse that doesn't just apply to these other kinds of normative discourse that set up the behavioral regulations for clubs and classrooms and so on? Well, I think we can, and I think we glimpsed it already. That is, moral discourse is a way of stepping outside of all of these local forms of regulation to aim to regulate human behavior more generally, apart from these local systems and their needs and demands and, frankly, functions. Moral discourse, one thing it does is enable us to always step back from any local system of rules, from the rules of this classroom to the laws of this country, and ask, sure, that's what is required in this classroom, but is that just? Should those forms of, say, dress be prohibited in an elementary school? Should those forms of speech be outlawed on college campuses? Should these kinds of laws be permitted in a modern democracy or anywhere, for that matter?
▶︎ 26:22 So moral discourse, when we can just ask, "Should it be this way?" enables us to step back from any other form of regulation and evaluate it and ask whether these are the regulations we ought to have and enable us then to give reason for revising our local standards or regulations. And we need this ability all the time. If nothing else, our circumstances change in ways that require us to review the ways we have previously regulated student behavior or whatever the case may be. So properly moral language, I think, at least one thing it does is to give us tools to try to publicly evaluate other systems of norms and regulations and publicly resolve disagreements about what these may be in ways that are supposed to be universal, that don't just appeal to my self-interest or that of my family or cronies or even not just to what my religion says, but that ought to apply to everyone, regardless of their interests, regardless of their local interests or culture or religion, without these dictates depending on anyone's attitudes so that we could all be wrong about that.
▶︎ 27:39 We have to discuss together what we really should adopt as behavioral standards or as regulations. And it's in this way that fully developed moral language can enable us to do very important things to being able to live together. We enable a kind of discourse where we can give shared permissions and requirements and also can justify them and step back to evaluate them in ways that aren't just what do I like or what benefits me. We can engage in a kind of generalized moral theorizing. We can ask questions about whether the morality of an action always depends on its consequences or not. Should we accept forms of utilitarianism?
▶︎ 28:25 And also, since we can reason, we can apply it to new situations because, again, our context is always changing. So if we have moral requirements against invasions of privacy, how do those apply to systems of digital surveillance, for example? We've got to ask that, and that wasn't part of what was originally in our practices. How should we apply these kinds of moral claims to our new situation? Or we have practices of holding people responsible if they hit someone with a car. How should this apply to driverless cars now and so on?
▶︎ 28:57 And as I've been emphasizing, we can not only apply our old system to new situations because we can reason with it, because we are able to put these regulations in the indicative and justify them in ways that don't depend on anyone's particular culture, interests, et cetera. We can always step back and evaluate, reevaluate, or revise any system of standards, again, in this kind of user-neutral way. I think if it's not user-neutral in this way, it's not a properly moral claim.
▶︎ 29:27 In this way, by taking us step by step on a functional story, by showing what each of these conceptual and linguistic developments, from imperatives to modal statements to nominalizations, what they each add to our ability to do stuff with normative discourse generally and moral discourse more specifically. By doing this, I think we can preserve what you might think of as the grain of truth in each of the prior apparently competing accounts of the function of moral discourse. As I mentioned at the beginning of this talk, there have been several different views.
▶︎ 30:06 Prescriptivists said, "Look, moral discourse basically functions to give commands, to prescribe forms of behavior." Well, that's absolutely true. That's the basic original form of it. That can be given in imperatives. But that's not all that sophisticated moral discourse does. And if you stop there, then you don't understand our ability to reason morally, for example, or to consider alterations to our moral language or to make generalizations.
▶︎ 30:38 Expressivists said moral claims like "hitting is wrong" serve to express our attitudes. It's like saying boo to hitting. Yeah, absolutely. That's another thing that the basic mood form of moral discourse can do is to express negative or positive attitudes towards different forms of behavior. That's cool. That's important, too. But again, if you stop there, you miss what else we can do with more sophisticated forms of moral discourse, such as engaging in moral reasoning, moral theorizing, and so on.
▶︎ 31:17 With descriptivists who wanted to say, "Look, moral discourse just tells you truths about the moral facts and properties," I want to say yes. That's part of the importance of getting it initially into the modal form, where you can put it in an indicative modal form. Then you can communicate a propositional content with this. If you say, "One mustn't hit," you can say, well, that expresses a proposition that one mustn't hit, and from that, maybe it follows that certain forms of physical abuse are wrong or something like that. That's all fine. It enables us to reason. That's really important.
▶︎ 31:55 But the thing to notice against what usually goes with this descriptivism is that that's not to say that this kind of talk aims to track a certain kind of moral fact or property in the way that talk of dogs or clouds enables us to track features of our environment. Instead, this kind of discourse gets introduced by introducing nominalizations, now not even directly out of observational discourse, but out of modal discourse. This is what the linguist Miriam Taverniers calls interpersonal grammatical metaphors rather than observational grammatical metaphors, which are like the property of redness. So we can move from "one mustn't hit one's brother" to "it is a requirement to not hit." And that introduces a new nominalization that enables us to do new things, but that's the way we get to it, not by anything remotely analogous to observing platypuses to introduce the term platypus. We introduce it by these licensed grammatical transfers that are part of the language for really good functional reasons.
▶︎ 33:07 In this way, we can explain why we have these kinds of nouns that have led to all the problems of metaethics. What are moral requirements? What are moral facts? What are moral properties? We can explain why we have this kind of discourse without saying we have it to track these things out there in the world. We can show how we could introduce it and why we would do so for functional reasons.
▶︎ 33:28 And there's nothing more we need. So we can get to what I've elsewhere called a kind of simple realism about moral facts and properties. Are there moral facts and properties? Yeah, sure there are. At least if you will grant me this much, if you will start from, if you think there are certain basic moral statements like one mustn't torture babies that are true, then you can make a trivial inference from that and say it is a moral fact that one mustn't torture babies. And the torturing of babies has the property of being morally wrong. And then we can say there are moral facts and there are moral properties in the only sense that fact and property talk has.
▶︎ 34:15 But this is a sense that's introduced by this process of nominalization. It's not something where we need to make an additional intuition or observation of something abstract or otherworldly to get this. So on this view, we can totally say there are moral facts and there are moral properties. Again, provided at least that you'll accept that, say, it's true that one mustn't torture babies or any other basic moral truth.
▶︎ 34:47 But the license to say this follows out of the grammatical rules that entitle us to introduce this terminology. It's not a case where we're positing moral facts or properties to try to explain some feature of our experience or discourse. So it's not taking that neo-Quinean route at all. But it is to say, yeah, sure, there's moral properties and moral facts and moral requirements in the only sense these terms have.
▶︎ 35:11 All right, so what does all this have to do with metaphysics? You might say, well, look, if you say there are moral facts and properties, then you haven't gotten us anywhere because all the same old metaphysical problems about what these things are arise again. Well, not so fast. For starters, as I just mentioned, we can easily resolve questions about whether there are moral facts and properties by these kinds of trivial inferences that introduce the discourse. We can also show why a lot of the problems that were supposed to plague those who said there are moral facts and properties and requirements and obligations and so on, why those problems arose from a faulty starting point, from a category mistake.
▶︎ 36:04 Some questions of metaethics include questions like, what sorts of things are moral requirements? Aren't these weird? Wouldn't they be spooky? How could they fit into the natural world? How could we have ever evolved a capacity for tracking them and coming to know about them? On the surface, these sound like perfectly reasonable, everyday questions. When I moved into my new house, there was something strange growing on the side of the tree. Turned out it was a bracket fungus. I didn't have that term before. Having been introduced to one, I might have questions. What sorts of things are bracket fungi? How do they fit into the natural world? Do they depend on the tree? Are they going to hurt the tree? All perfectly good questions.
▶︎ 36:47 But the term moral requirement isn't like the term bracket fungus, and the fact that those questions make sense for bracket fungus doesn't tell us that those questions make sense for moral requirements. The term bracket fungus is introduced observationally. You see something in the environment, you learn the word from someone who knows it then, and it functions to track observable features so you can find out if it's gotten bigger or smaller, if there's others in your forest, and so on. But the term moral requirement isn't like that. It's not introduced observationally. It's introduced by introducing a new nominalization out of more basic inferences from modal claims. From one mustn't to there's a prohibition, for example.
▶︎ 37:38 And what it does for us isn't enables us to track the moral requirements. Instead, it enables us to do useful generalizing things, to qualify and quantify the requirements, to form moral generalizations, to address theoretic questions in moral philosophy about the relationship between the moral requirements and appropriate punishments or the tendency to cause pain or whatever. So the classic metaphysical questions that lead people to be suspicious of the idea that there are or really are moral facts and properties arise from false analogies with the kind of talk that we learn first, observational talk. You might even think of them as a category mistake. That is, a mistake that arises from a false presupposition about what this discourse is like and how it works.
▶︎ 38:33 So if you use a noun term like requirement, that sounds superficially like other nouns, like bracket fungus. And so that might trigger a presupposition that we're naming a thing, a thing that we ought to be able to observe and track and say more about, and that where this term ought to be observationally acquired. But that's not how terms like moral requirement enter language, and it's not how they work. So as a result, we can see that some of the problems that have been raised and discussed a whole lot in metaethics are really wrong-headed and arise from a false presupposition about how this language works.
▶︎ 39:19 Questions like, how can these things be placed in the natural world? They're not supposed to be placed in the natural world. This is a noun term introduced for new theoretic and generalizing purposes. Are moral properties grounded in natural properties, and if so, how? What are they based on? As if this is like the question of how anemia is based in changes at the cellular level. Cool question for anemia, not great for moral requirements.
▶︎ 39:46 How can moral facts be action-guiding? And people who talk about this problem talk of it as if there's some mere thing in the world that should be pushing us to action. But that's not the case. Instead, these forms of moral discourse develop and evolve from what would originally be expressed as imperatives, which clearly have an action-guiding function, and then they just get these more grammatically sophisticated forms that enable us to do other things at the same time. Or complaints that we couldn't possibly come to know moral facts if we aren't causally connected to them because we couldn't have evolved a capacity for reliably tracking them. Well, wait, tracking's the wrong model for moral requirements, moral properties, moral facts to begin with. Tracking's a model that works great for observationally acquired terms and the things that we're referring to with them. But we shouldn't have ever expected moral knowledge to go that way.
▶︎ 40:41 Okay, so why is the linguistic story relevant to those who are interested in the metaphysical questions of metaethics? Well, as I suggested, it means that some of the old metaphysical questions like are there moral facts or properties, are easily answerable. Yes, there are, in the only sense that has sense, but no, they're not explanatory posits. Other metaphysical problems or alleged metaphysical problems can be seen as arising from mistaken presuppositions about how the language works, a functional monist assumption. And if that's the case, then our best move is to expose the mistaken presupposition, not to try to answer the question straightforwardly and show that moral facts are somehow grounded in features of the natural world.
▶︎ 41:28 Where does this leave metaethics? If I'm right, this gives us the basis, at least, for rethinking a lot of classic metaphysical problems in metaethics. Doesn't mean we're all done. There's still all sorts of interesting philosophical questions in the vicinity. There's, of course, loads of first-order moral problems. What should we do type problems. What kinds of action should we prohibit or should we engage in? What makes a virtuous life, et cetera. I haven't touched on any of those.
▶︎ 41:56 Although one thing we've said is that cohesively with the function of properly moral discourse, any real moral justification for a claim has got to be impersonal, not based on how much it will benefit me, not based on what I or my comrades believe or want. It's got to be neutral like that to count as properly moral at all. So we can give room for rejecting some claims about what we should or shouldn't do, or at least rejecting the idea that these are moral.
▶︎ 42:25 It'll also leave us with certain constructive and theoretic problems. We shouldn't expect that we gain moral knowledge by having evolved a reliable capacity for tracking moral facts or anything else. How can we gain moral knowledge? How can we also systematize moral claims? Or can we, or is there no such system? Are there general moral theories available, whether it's a utilitarian approach or a Kantian approach or whatever? So there's all sorts of interesting theoretic questions and moral questions still available.
▶︎ 42:58 What I think we get a new view on in a way that enables us to resolve them or relieve ourselves of them is a lot of the metaphysical questions of metaethics. And I've tried to show that doing this enables us to capture the element of truth in prior theories, including the prescriptivist, expressivist, and even the descriptivist approach, and show that they're each onto something about the functions of some forms of moral discourse. And we can get a more complete theory that involves all of those in a story about how these different forms of moral discourse develop and why, why we would want them, what functions they serve for us.
▶︎ 43:33 And then I've tried to show how that enables us to reevaluate some of the classic metaphysical problems of metaethics and even help us to see how a view like mine can understand and address and work on practical questions about what we should do without thinking of that as we're discovering some metaphysical thingies out there, the moral facts and properties. So we can engage in the kind of practical questions that's needed for work in conceptual engineering without having to do serious old-fashioned metaphysics that thinks of that as discovering deep, hidden truths about the world.
▶︎ 44:17 In doing this, I've aimed to then also, you might say, deflate some of the metaphysical questions and problems in metaethics. Let me just close by saying why I think this is important. If you think about the problems of morality in a metaphysically serious way, where you think that in order for moral claims to be true, there must be these kinds of hidden special kinds of moral properties and requirements that aren't observable, that we can't interact with, that wouldn't be part of our natural world, and you think of these as their own special metaphysical kind of thing. That might sound good for a while, but there's always going to be people who have a very skeptical attitude towards this, who can't make sense of the idea of these non-natural facts or properties out there in the world that we could somehow come to know through intuition or something. And that can lead to forms of moral skepticism and even moral nihilism.
▶︎ 45:27 If you can see what's going on with moral discourse in the more down-to-earth way that I've been developing it, we can see why it's really important and why there are standards and why you can't just say anything you like, but without thinking of it as requiring a positing of some mystical metaphysical entities. And if you can do that, then you can also fend off some of the motivations for skepticism and nihilism in a way that I think is really important.
▶︎ 45:58 So this is just one example of the way in which taking seriously the different kinds of functions that language can serve, and making use of work in linguistics that can show us how these functions get added up through more and more sophisticated forms of discourse. How this can help us take a new view of old problems in metaphysics, answering some easily, unraveling others as based in mistaken presuppositions. And this is all part of rethinking metaphysics. In the next lecture, I'll try to do something like this approach also to help us understand better the old problems of truth.