Options
Article

After Metaethics

Authors
  • Tristram McPherson (The Ohio State University)
  • David Plunkett (Dartmouth College)

Abstract

In recent work, several philosophers have begun to explicitly explore the conceptual ethics of normativity. Put roughly, this is a kind of normative and evaluative inquiry that aims to assess the normative words and concepts that we currently use, as well as salient possible alternative normative words and concepts that we might choose to adopt. One important question about this project is how it relates to more familiar metaethical or metanormative inquiries. This paper helps to illustrate the deep ways in which metanormative and conceptual ethics inquiries can interact, and why it is valuable for practitioners of these projects to attend to the relationships between them. The paper begins by introducing a powerful (although not exhaustive) model for thinking about this interaction, which generalizes on the recent “after error theory” literature. We argue that this model allows us to identify important and neglected views about the normative, and to better understand the argumentative burdens that philosophers take on in defending such views. We conclude by briefly discussing another way that thinking about error theory can be important for understanding the relationship between the conceptual ethics of normativity and metanormative inquiry: by complicating how we can understand the project of conceptual ethics itself.

Keywords: conceptual ethics, conceptual engineering, metaethics, metanormative inquiry, error theory, abandonment, eliminativism, conceptual reform

How to Cite:

McPherson, T. & Plunkett, D., (2025) “After Metaethics”, Philosophers' Imprint 25: 23. doi: https://doi.org/10.3998/phimp.4357

Downloads:
Download PDF
View PDF
Download XML

212 Views

26 Downloads

Published on
2025-10-27

Peer Reviewed

License

Creative Commons Attribution-NonCommercial-NoDerivs 4.0

Introduction

Recently, several philosophers have begun to explicitly reflect on the conceptual ethics of normativity as a distinctive part of normative theory.1 Put roughly, the conceptual ethics of normativity aims to assess the normative words and concepts that we use, as well as salient possible alternative ones. One important question about the conceptual ethics of normativity is how it relates to more familiar metanormative and metaethical inquiry. Much discussion in ethical theory skates over or obscures systematic questions about the relationship between these projects. For instance, Peter Railton’s “reforming” account of “moral goodness” involves a conceptual ethics proposal about how to amend an important ethical concept.2 And yet, it is commonly treated as a competitor to metaethical proposals with no “reforming” aspects, which (at least prima facie) seem to have different explanatory aims.

In this paper, we explore the relationship between metanormative inquiry and the conceptual ethics of normativity, and why it matters for our thinking about the foundations of ethics. To do so, we begin with an entry point that we take both to be illuminating and to be comparatively familiar for those working in metaethics: metaethical error theory and the “after error” question it prompts. We use this entry point to motivate a general (and, we argue, explanatorily powerful) framework for understanding important ways in which work in metaethics and conceptual ethics can interact. We argue that this framework helps illuminate how we should best understand a range of important existing claims about the foundations of ethics, as well as possible new ones, and, moreover, how to best argue for those claims.

We proceed as follows. In §1, we briefly introduce a familiar form of metaethical error theory, and discuss how it has been used to motivate a range of “after error” projects. In §2, we more carefully distinguish metanormative inquiry from the conceptual ethics of normativity, and put this distinction to use in organizing our thinking about the “after error” literature. Then, in §3, we show how generalizing from this discussion provides a fruitful (although not exhaustive) model for understanding the interaction between inquiries in metaethics and a salient branch of the conceptual ethics of normativity: the conceptual ethics of ethics. We call this the After Metaethics model.

We use this model to show that a range of prominent arguments in the metaethics literature, about positions other than error theory, can also be used to motivate projects in the conceptual ethics of normativity. We illustrate our case with three central examples: Christine Korsgaard’s argument against normative “realism,” David Enoch’s “schmagency” objection to constitutivism, and objections to metaethical subjectivism based on issues about disagreement and non-arbitrariness. Each of these examples, we argue, invites a neglected and philosophically fruitful “conceptual ethics” reading. We then briefly emphasize the diversity of interesting “after metaethics” projects. Specifically, we illustrate the range of (purported) desiderata that one might appeal to and the range of conceptual ethics positions one might argue for, in such projects.

Finally, in §4, we consider another important kind of interaction between metanormative inquiry and conceptual ethics. This concerns the implications of metanormative inquiry for engaging in conceptual ethics inquiry. We argue that there are strong motivations to appeal to “authoritative” norms in conceptual ethics inquiry. However, some forms of metanormative error theory deny the existence of authoritatively normative facts. That is: they deny the existence of the very normative facts that (we claim) it (at least prima facie) makes sense to appeal to in doing conceptual ethics. Indeed, in some cases, metanormative error theorists might deny the existence of any normative facts whatsoever. This raises difficult questions in the foundations of conceptual ethics, including ones about what kinds of normative claims (if any) conceptual ethics inquiry can really support.

1. Error Theory and After

In this section, we introduce a canonical form of error theory in metaethics, and the “after error” question that this theory naturally prompts.3

We can orient to the canonical error theorist’s basic view about ethical thought and talk by noting that it is akin to a familiar sort of atheist’s view about theological thought and talk. According to this sort of atheist, standard theological thought and talk purports to be about God, but there is no God. So, according to this atheist, standard theological thought and talk is shot through with a fundamental error. Similarly, the canonical error theorist thinks that standard ethical thought and talk enshrines a fundamental error.

More specifically, canonical error theory is characterized by a commitment to the following three ideas about ethical thought, talk, and reality.4 First, at the level of thought, canonical error theorists are cognitivists: they claim that at the most basic explanatory level, ethical thought consists in ethical beliefs. Second, at the level of talk, canonical error theorists are descriptivists: they claim that declarative ethical sentences purport to describe the world. The unifying idea in both cases is that ethical thought and talk serves to represent the world as having ethical features. Thus, on this picture, when Anja thinks that torture is wrong, her thought represents acts of torture as instantiating an ethical property (namely: wrongness). Finally, at the level of reality, canonical error theorists are ethical nihilists. That is, they think there are no ethical facts or instantiated ethical properties in reality. Typically, this is because the error theorist thinks that—in order to satisfy our ethical concepts—ethical facts would have to have certain features (such as being “objectively prescriptive,” “irreducibly normative,” or providing us with “categorical reasons for action”).5 But, according to the error theorist, nothing in reality possesses the relevant combination of features.

These three commitments entail that our ethical thought and talk exhibits the error that gives canonical error theory its name: our ethical thought and talk, while purporting to represent ethical aspects of reality, systematically fails to do so, because there are no such things for it to represent. Consequently, ethical thought and talk involves a kind of delusion about the way reality really is.

Accepting canonical error theory has tended to prompt what we will call the “after error” question: namely, what should we do with our ethical thought and talk going forward?6 Many philosophers find this question to be more pressing in the case of error theory about ethics than in the case of many other error theories (e.g., about phlogiston, which some people once falsely believed was a kind of substance released in combustion). Why is this? We suspect that the answer has something to do with the seeming centrality and importance of ethical thought and talk in so much of our lives (where this contrasts, for example, with thought and talk about phlogiston). For our purposes here, however, what matters is not why many drawn to metaethical error theory find this to be such a pressing question. Instead, what matters is what philosophers have said in response to it. The core options discussed in the literature are as follows.7

A first option is to abandon (or “eliminate”) the use of ethical thought and talk (either altogether, or rather just in some important range of contexts).8 The desire to avoid the systematic error alleged by the canonical error theorist provides an apparently powerful consideration in favor of abandonment.9

A second option is to retain our ethical thought and talk, despite the errors it (allegedly) involves.10 A prominent motivation for this kind of view appeals to the important roles ethical thought and talk can play in our lives in many contexts. These include, for instance, guiding deliberation and coordinating activity. Different versions of this motivation can support a range of views about how much of ethical thought and talk we should retain, and in what sorts of contexts.11

A third option is to reform our existing ethical thought and talk in some way, or to replace it with some alternative form of thought and talk. This option is motivated by the idea that ethical thought and talk could be amended such that it can continue to play the sort of important roles in our lives mentioned above, without involving the errors that it allegedly involves. (Whether the relevant amendments amount to “reform” or “replacement” is a delicate matter that we will not delve into here). Two notable examples of such proposals are “revolutionary fictionalism” and “revolutionary expressivism.”12

In our view, the “after error” literature helpfully exemplifies a striking kind of relationship between metaethics and the conceptual ethics of ethics. As is conventional, we take canonical error theory to be a contribution to metaethical inquiry. By contrast, we take the “after error” question to be a question in the conceptual ethics of ethics, because it asks a normative question about our ethical thought and talk. So we appear to have a metaethical conclusion motivating a conceptual ethics inquiry.

The intimate relationship between error theory and the “after error” question, however, might prompt a different reaction: a suspicion that there isn’t an interesting distinction between metaethics and the conceptual ethics of ethics. This suspicion might be bolstered by the observation that many philosophers tend to classify the “after error” literature as a part of metaethics. We think that suspicion tracks something important about how many people currently use the term ‘metaethics’.13 However, in the next section, we argue that characterizations of “metaethics” and the “conceptual ethics of ethics” that distinguish these projects can be well motivated, despite the intimate connections that are possible between them.

2. Metaethics and the Conceptual Ethics of Ethics

As with many philosophical terms and phrases, ‘metaethics’ and ‘the conceptual ethics of ethics’ are each used in a variety of ways. Our aim in this section is not to offer characterizations of “metaethics” and “the conceptual ethics of ethics” which capture every such use. Rather, we aim to offer characterizations that pick out distinctive and philosophically important sorts of inquiry, which we take to answer central interests inquirers have in using these labels as well as important strands of the history of how these labels have been used. The characterizations we offer here stem from our previous work, in which we explain, motivate, and defend these characterizations at more length than we have space to do here.

We begin with metanormative inquiry, of which we take metaethics to be a subpart. We characterize metanormative inquiry as follows:

Metanormative Inquiry Metanormative inquiry aims to explain how actual normative thought and talk—and what (if anything) that thought and talk is distinctively about—fits into reality.

Here we will very briefly unpack a couple of key elements of this characterization.14

First, for brevity, here and elsewhere in this paper, we use the word ‘normative’ broadly, to encompass both the normative and the evaluative. Second, we understand the notion of what certain thought and talk is “about” here broadly, in an “intensional” sense. For example, in this sense, the word ‘phlogiston’ is about something (namely, a certain kind of substance released in combustion), even though there is no such substance. Third, on this account, metanormative inquiry is a kind of descriptive inquiry. It aims to correctly describe (and explain) actual normative thought, talk, (perhaps) things such as normative facts, properties, and relations, and how all of these fit into reality. Finally, different theorists have different accounts of what “reality” amounts to. Here, we take reality to be the totality of what there is.

We understand metaethics to be a branch of metanormative inquiry that takes up the relevant explanatory project specifically concerning ethical thought and talk, and what (if anything) that thought and talk is distinctively about. (Other branches of metanormative inquiry include, for example, metalegal and metaepistemic inquiry.) Familiar metaethical questions, such as “are ethical judgments really beliefs or are they more like desires?” and “are there instantiated ethical properties?”, naturally arise within this explanatory project. In most contexts, it makes sense to classify canonical error theory as a “metaethical theory,” given its ties to this project. In short, this is because by offering answers to such questions, it provides the outlines of a comprehensive answer to the explanatory question that is constitutive of metaethical inquiry.

Now we turn to our account of the conceptual ethics of normativity, which draws on the account of “conceptual ethics” that one of us (Plunkett) first developed with Alexis Burgess.15 As we understand it, “conceptual ethics” is inquiry into a cluster of related normative and evaluative questions about thought and talk. There is no theory-neutral way to state what exactly that cluster includes. However, for our purposes here, we can say that it includes questions about the normative and evaluative assessment of concepts and their use, as well as parallel questions about words and about concept-word pairings.

We understand the project of conceptual ethics broadly, in the following two ways. First, as suggested just above, work in conceptual ethics need not be about concepts. For example, one could engage in normative or evaluative inquiry concerning the pairings of words with semantic values, inferential roles, and pragmatic phenomena. Second, work in conceptual ethics need not be tied to distinctively ethical norms. For example, such work could assess concepts in terms of non-ethical standards such as how well they facilitate inquiry or carve reality at its joints. The breadth of this gloss makes the label ‘conceptual ethics’ potentially misleading. However, we stick with it both because we aren’t convinced that there is a better shorthand label for the relevant sort of inquiry, and for continuity with the existing literature.16

In parallel to treating metaethics as a subset of metanormative inquiry, we treat the conceptual ethics of normativity as the subset of conceptual ethics which is focused on normative and evaluative thought and talk. In turn, we take the conceptual ethics of ethics to be a further subset of the conceptual ethics of normativity, focused on ethical thought and talk in particular.17 With this gloss in hand, we can see that the “after error” question—Given error theory, what should we do with our ethical thought and talk, going forward?—is a question in the conceptual ethics of ethics: it is a normative question concerning whether to (e.g.) retain, reform, replace, or abandon our ethical thought and talk.

These accounts of metanormative inquiry and the conceptual ethics of normativity allow us to highlight two of the most important contrasts between these projects.

First, as we emphasized above, metanormative inquiry is a kind of descriptive inquiry. Thus, it aims to accurately explain (for example) how our actual normative thought and talk in fact works, not to evaluate such thought and talk. In contrast, the conceptual ethics of normativity is a kind of normative inquiry. For example, it concerns questions about the kind of normative thought and talk we should engage in, or about which kinds are better or worse.

Second, metanormative inquiry concerns actual normative thought and talk (and what, if anything, it is distinctively about). By this, we mean that it concerns thought and talk using the actual normative words, concepts, etc. that people use (or have used in the past) in the actual world. In contrast, the conceptual ethics of normativity has a wider scope. Some work in conceptual ethics focuses on the defects or virtues of our actual thought and talk. However, conceptual ethics is also often concerned with the evaluation of possible salient alternatives to that thought and talk. To help us focus on some of the most relevant salient alternatives, we use the term ethical-ish to talk about possible forms of thought and talk that share many of the core inferential, representational, and communicative aspects of actual ethical thought and talk.

The two contrasts we have just explained constitute theoretically deep differences between metanormative inquiry and the conceptual ethics of normativity. However, this doesn’t mean that these projects can’t interact in interesting ways. They indeed can. For example, a given claim (or theory, question, etc.) can figure prominently in both of these projects.18 We take this to be the case for the canonical error theory in metaethics that was discussed in the preceding section. That example illustrates both the contrast between these projects, and certain interesting relationships that can arise between them.

3. Generalizing: After Metaethics

In this section, we show that we can abstract from the “after error” literature to provide a fruitful general model for thinking about a range of important interactions between metaethics and the conceptual ethics of ethics. We then put this model to use, showing how it illuminates issues raised by three prominent arguments in contemporary metaethics. Finally, we step back from these examples to illustrate the diversity of interesting projects helpfully understood in terms of the After Metaethics model.

3.1 The After Metaethics Model

We begin by explaining how the structure of “after error” inquiry suggests an important (although not exhaustive) general model for how metaethics and the conceptual ethics of ethics can interact. This model begins with a kind of conceptual ethics inquiry that seeks to identify desiderata for ethical-ish thought and talk. As we understand them, “desiderata” are things that it would be good for something (e.g., ethical-ish thought and talk) to have. In the “after error” literature, the (purported) desiderata characteristically include avoiding reference failure, presupposition failure, or commitment to falsehoods in one’s thought and talk. Note two central questions here: first, what the genuine desiderata for a given fragment of thought and talk actually are; and, second, how to weigh those desiderata against each other. (As our use of the phrase ‘purported’ above suggests, we take it that people can be mistaken in their beliefs about both of these things. In what follows, we often drop the “purported” modifier for ease of presentation).

Once one has identified such desiderata, one might ask either of two different questions. First, one further sort of conceptual ethics project would be to ask: what possible sorts of ethical-ish thought and talk would secure the identified desiderata? (We can call this possible thought and talk inquiry.) Reform or replacement options in the “after error” literature involve this latter sort of inquiry, and can thus be used to orient to what it involves. Second, it is often natural to want to know specifically whether actual ethical thought and talk satisfies the relevant desiderata. The next step in our model is to engage in metaethical inquiry to answer this question. (We can call this actual thought and talk inquiry.) If our actual ethical thought and talk does satisfy the desiderata, this would provide a conceptual ethics vindication of our existing ethical thought and talk, relative to those desiderata. In our example, however, the canonical error theorist concludes that the relevant desiderata are not satisfied. This naturally motivates further conceptual ethics inquiry: given that our actual ethical thought and talk (allegedly) fails to meet the relevant desiderata, what should we do? In our example, this was illustrated by debates in the “after error” literature, concerning whether we should abandon, retain, or reform/replace actual ethical concepts. We can generalize from this case, to frame a broader after metaethics project, which is motivated by the relationship between metaethical inquiry and desiderata on our ethical thought and talk. In this general case, the key post-error theory options—to abandon, retain, or reform/replace our ethical thought and talk—remain the salient ones.

We can sum up what we will call this After Metaethics model in the following diagram.

Here is how we think of this model as illustrating a central form of interaction between inquiries in metaethics and the conceptual ethics of ethics. First consider actual thought and talk inquiry. This involves inquiry into a paradigmatic kind of metaethical question: namely, the descriptive question of what our actual thought and talk is like. Within the model, however, this inquiry is also playing a conceptual ethics role, as it concerns the evaluation of that thought and talk. The rest of the boxes in our model involve familiar kinds of conceptual ethics inquiry or conclusions. The model thus illustrates how work in conceptual ethics can both motivate, and be motivated by, work in metaethics. The after metaethics box, in particular, marks a kind of conceptual ethics project paradigmatically motivated by the results of metaethical inquiry.

It is important to emphasize three things about the After Metaethics model.

First, we take this model to provide a theoretically illuminating way of understanding a wide range of ways that inquiries in metaethics and the conceptual ethics of ethics can interact. However, we do not take it to be an exhaustive account of possible interactions between these sorts of inquiry.19

Second, this model is intended as an analytical structure, rather than as a prescription for a process. For example, some inquirers might first accept a metaethical view, and only then begin thinking about desiderata for ethical-ish thought and talk. Or they might revise their views about such desiderata on the basis of their informed metaethical views (e.g., holding that the relevant notion of “reference” they initially had in mind was too demanding).

Third, there are two places that desiderata for ethical-ish thought and talk can enter the model. Most obviously, they enter into the model at the “desiderata thesis” stage. However, such desiderata are also crucial at the “after metaethics” stage. As the example of canonical error theory illustrates, some projects are framed around testing ethical-ish thought and talk for certain important desiderata. However, at the “after metaethics” stage, we may appeal to a wider set of desiderata, in adjudicating among various versions of abandonment, retention, and reform/replacement.

The After Metaethics model can help explain a striking contrast between the philosophical literature about metaethical error theory and the literature about other prominent metaethical views. As we have seen, in the case of error theory, discussions of abandonment, retention, or reform/replacement are common. And even when authors don’t advance a view on that question in a given discussion, they often bring it up as an important question for future discussion.20 That is, discussions of what we call after metaethics projects are highly salient in this literature. By contrast, there isn’t a comparable literature on the questions “after non-naturalistic realism, what?” or “after expressivism, what?” Such discussion does exist.21 But in cases other than error theory, it is much more common for systematic discussions of metaethical views to not even bring up the question “after view X, what?”

We think that there is a clear explanation for this difference. Consider two facts about the desideratum of avoiding widespread error in ethical thought and talk. First, it is relatively uncontroversial that this is a genuine desideratum on ethical-ish thought and talk. Second, it is relatively transparent that if canonical error theory is true, this desideratum is not secured by our actual ethical thought and talk. By contrast, contemporary metaethical theories other than error theory are typically presented by their proponents as vindicating the core commitments of our actual ethical thought and talk, and it is often seen as a goal for such theories that they do so.22 Because these theories are well understood as implicitly committing to the Vindicatory conclusion outcome within the After Metaethics model, these theories, as presented, do not motivate engagement in the after metaethics project.

3.2 “Conceptual ethics” readings of three prominent arguments in metaethics

While the reasoning just given explains the lack of attention to “after metaethics” projects in non-error-theoretic metaethics, we do not think it vindicates it. In this section, we illustrate the interest of after metaethics projects by discussing three arguments that have loomed large in contemporary metaethics. We suggest that there are natural readings of these arguments on which their primary conclusions are in conceptual ethics, rather than metaethics.23 And understood in this way, these arguments can be used to motivate neglected after metaethics projects. It is worth emphasizing that our aim in this section isn’t exegetical. Rather, it is to suggest philosophically rich ways of thinking about and developing important kinds of arguments.

First, consider Christine Korsgaard’s objections to the view she calls “realism” in The Sources of Normativity.24 Put roughly, for Korsgaard, “realism” about normativity involves the idea that there are normative facts that are independent of facts about how agents engage with reality from the “practical point of view.”25 At the core of Korsgaard’s objection to this kind of theory is her idea that agents confront what she calls “the normative question.” Roughly, this means that they will encounter moments when doing what morality requires is hard, which can naturally lead them to ask why they should do what it prescribes. Korsgaard thinks agents should want an answer that makes morality’s demands intelligible to them from their own point of view.26 Famously, Korsgaard charges that “realism” cannot provide such an answer.27

Next consider David Enoch’s “schmagency” objection to constitutivist metaethical theories. (Korsgaard’s own “Kantian constructivism” is one such constitutivist theory). Put roughly, the kind of “constitutivist” accounts Enoch targets claim that the fundamental normative facts are explained by facts about what is constitutive of agency. As Enoch frames the issue, constitutivism is motivated by the attempt to (i) connect normativity to our psychology while (ii) doing so in a way that is not objectionably arbitrary. Enoch argues that constitutivist accounts remain arbitrary in the relevant sense: even if (for example) I am committed qua agent to doing what morality requires, I can sensibly ask “so what?” because agents “need not care about their qualifications as agents.”28 Enoch bills this objection as showing that constitutivist accounts cannot “ground” normativity.29 Based on this, he objects to such accounts as metaethical views.30

Third, consider “subjectivist” metaethical views, according to which, put roughly, normative claims describe the psychological states (e.g., the desires, or other non-cognitive attitudes) of the speaker making those claims. Such views, it is often claimed, would render genuine or “substantive” ethical disagreement impossible. This impossibility, in turn, is often taken to constitute a powerful objection to subjectivist metaethical views.31

In each of these cases, it is alleged that certain problematic consequences would result if a certain metaethical view were correct. And this allegation is used in turn to object to that metaethical view. We grant that it is possible to argue in this way. However, we will use our After Metaethics model to argue that in each of these examples, the alleged problematic consequence more directly motivates a conclusion in the conceptual ethics of ethics, rather than a view in metaethics. We think that the conceptual ethics versions of these arguments are underexplored, and well worth taking seriously. Indeed, in some cases, we think the arguments might in fact best be developed in this way.

Within our After Metaethics model, we can treat each of the cases discussed as beginning by identifying an (alleged) desideratum on ethical-ish thought and talk. Respectively, these desiderata are: enabling agents to answer “the normative question”; avoiding arbitrariness; and enabling genuine disagreement. The arguments can then be seen as claiming that a certain view (“realism,” constitutivism, and subjectivism, respectively) fails to meet that desideratum. What sort of conclusion can we draw at this stage? Most naturally, we can think of these arguments, so read, as engaging in the “Desiderata” and “Possible thought and talk inquiry” stages in our model. The conclusions drawn are ones in conceptual ethics, which need not involve a claim about how our actual ethical thought and talk work.

To drive home this point, consider in more detail our “Korsgaardian” reasoning (again, we emphasize that our main concern here is not whether this precisely captures Korsgaard’s actual arguments):

  • P1. Agents confront the normative question (roughly, the question of what to do in hard cases).

  • P2. Because of this, in deliberation, agents ought to use ethical-ish concepts that enable them to answer the normative question in a way that is intelligible to them.

  • P3. If realism is true of our ethical concepts, then our ethical concepts cannot enable agents to intelligibly answer the normative question.

Now consider two conclusions that one might try to draw from these premises:

Korsgaardian metaethical thesis Realism is not true of our ethical concepts.

Korsgaardian conceptual ethics thesis If realism is true of our ethical concepts, then agents ought not to deliberate with ethical concepts.

One contrast between these conclusions is this: the Korsgaardian conceptual ethics thesis follows directly from the argument’s premises. By contrast, it is far from clear how (if at all), these premises support the Korsgaardian metaethical thesis.

As we see it, the Korsgaardian conceptual ethics thesis naturally motivates instances of the two sorts of inquiry that form the “middle layer” in our diagram:

  • (a) inquiry into what possible sorts of ethical-ish thought and talk would enable agents to answer the “normative question.”

  • (b) inquiry into whether our actual ethical thought and talk enables agents to answer the “normative question.”

In order to establish the Korsgaardian metaethical thesis, the Korsgaardian would have to engage in actual thought and talk inquiry. Nothing we are saying here is intended to rule out this possibility. If it can be established that our actual ethical thought and talk does enable us to answer the “normative question,” then this could be combined with P3, above, to argue that realism in metaethics is false. 32 Furthermore, consider that Korsgaard believes that only (Kantian) “constructivist” concepts can enable agents to answer the normative question.33 If so, showing that our actual ethical concepts enable us to answer this question would entail that Kantian constructivism (which, roughly, can be understood as a version of the sort of “constitutivist” view we glossed above) is the correct metaethical view.34

What we most want to emphasize is the following: Suppose one were fully persuaded by premises 1–3 above, but agnostic or pessimistic about the sort of additional metaethical argument just mentioned. In such a case, the Korsgaardian argument could still be extremely interesting because it helps establish an interesting conclusion in conceptual ethics.

To see this, consider again that Korsgaard’s core complaint is that realism fails to address the “normative question.” This complaint is separate from the question of whether “realism” is a good description of our actual ethical thought and talk. To press this point, suppose that our existing ethical thought and talk is “realist” in Korsgaard’s sense. If so, ethical thought and talk might fail to do the thing Korsgaard wants it to do: namely, enable agents to satisfactorily answer the “normative question.” This failure would naturally motivate an “after metaethics” project to reform or replace our existing ethical concepts, so that the ethical-ish concepts we use would help agents to successfully answer the normative question. That could motivate an argument in conceptual ethics for the thesis that we should adopt a “Kantian constructivist” form of ethical-ish thought and talk.35 In other words, it could be the start of an argument for what we might call “revolutionary Kantian constructivism.”36

Similar points hold for the other examples we introduced. Consider Enoch. Suppose that he is correct that constitutivist metaethical views fail to explain how agents could make non-arbitrary choices. And suppose further that only the sort of non-naturalistic realism he favors could do so. (Put roughly, on this sort of view, there are ethical facts that are fundamentally different in kind from the sort of “naturalistic” ones studied in the natural and social sciences, and our ethical thought and talk successfully refers to such facts.) This is all compatible with the possibility that constitutivism provides the correct account of our actual ethical thought and talk. In order to establish that non-naturalism is the correct metaethical view, Enoch needs to establish that our actual ethical concepts in fact enable us to make non-arbitrary choices.

But, again, suppose that we are pessimistic about whether our actual ethical concepts in fact enable us to secure the relevant desideratum (namely, of allowing us to make the relevant sorts of non-arbitrary choices). This is compatible with the possibility that there is a possible, ethical-ish form of thought and talk of which non-naturalism is true, which we could use to make relevantly “non-arbitrary” choices. And perhaps there really are the relevant sorts of “non-naturalist” ethical facts in the actual world for that thought and talk to successfully refer to, even though it currently does not do so.37 This might motivate advocating for “revolutionary non-naturalism” on the basis that this is the way of reforming our ethical thought and talk to enable it to provide a framework for non-arbitrary choice.

Consider another way that the possibility of revolutionary non-naturalism might be philosophically significant. Derek Parfit understood himself as having dedicated his life to investigating a kind of “irreducible” normativity. Notoriously, Parfit claimed that, if certain metaethical views (such as reductive naturalism) were true, then his life would thereby have been wasted.38 But suppose that revolutionary non-naturalism is the correct view in the conceptual ethics of ethics. Then Parfit’s life’s work might have been worthwhile, in virtue of helping us to understand the properties that we should be talking about with our ethical thought and talk, even if non-naturalism is not true of our actual ethical thought and talk.

Finally, consider subjectivism. As we noted above, some have argued that the subjectivist cannot explain the possibility of “substantive” ethical disagreements.39 We can distinguish two questions about such disagreements:

  • (i) Do our actual ethical concepts in fact enable such disagreements?

  • (ii) How important is it for ethical-ish thought and talk to enable such disagreements?

Suppose that we combine a positive answer to (ii) with a negative answer to (i). This suggests that subjectivism might well be true of our actual ethical thought and talk, but that this has regrettable implications. One could use these answers to argue for revolutionary invariantism, according to which the semantics of our ethical terms should be amended so that their truth conditions are insensitive to the ethical attitudes of conversational participants.

Suppose instead that we combined a positive answer to (i) with a negative answer to (ii). On this picture, our actual ethical thought and talk is invariantist, but it is not that important for this to be true. If we supposed that subjectivism had other desirable features (for example, perhaps the epistemology of a subjectivist view is especially straightforward), then we might, on this picture, consider advocating for revolutionary subjectivism.

We want to conclude this subsection by briefly emphasizing two important points about these arguments. The first concerns how developing these arguments in the ways we have suggested matters for the evidence we look to in assessing them. The second concerns the relationship between engaging in conceptual ethics and our ability to illuminate the ethical reality we have been studying prior to such engagement.

The point about evidence is relatively straightforward. As we have emphasized, conclusions in conceptual ethics are characteristically normative, while conclusions in metaethics are characteristically descriptive. Because of this, the evidence one needs to defend each kind of claims will be different. To illustrate, suppose that one developed the Korsgaardian argument above with the aim of establishing the Korsgaardian metaethical thesis we discussed. If so, then, given certain (we think quite reasonable) assumptions, empirical evidence from the natural and social sciences (e.g., linguistics, cognitive science, and psychology) would be an extremely rich source of evidence that has direct bearing on whether this conclusion is correct. In contrast, suppose one develops the Korsgaardian argument with the aim of establishing the Korsgaardian conceptual ethics thesis. Here, one’s arguments would need to appeal centrally to evidence for normative claims about which sorts of concepts we ought to use in contexts like deliberation. And it is famously much less clear that the sorts of empirical research mentioned above would be of central relevance to such claims. (We do not mean to overstate this contrast: it is a familiar thought, which we endorse, that empirical work can be deeply relevant to normative arguments in a variety of ways.)

Now turn to the second point. Suppose that, in light of a conceptual ethics argument, one proposes to reform or replace our ethical concepts with novel ethical-ish concepts. Can we use those reformed concepts to think and talk about, and investigate ethical reality? One might think not. This is because one might think that the term ‘ethical reality’ (insofar as it successfully refers) picks out that part of reality that constitutes the extensions of our actual ethical terms or concepts.40 If this is right, then if we reformed or replaced those concepts with ethical-ish concepts that have even slightly different extensions, one might think that we would then, strictly speaking, be thinking and talking about a different part of reality (if we manage to refer to any part of reality at all).

A theme in contemporary conceptual ethics, however, suggests that the above reasoning is too quick. Several philosophers have argued that some ways of reforming or replacing a concept can preserve the topic that was being addressed by the unreformed concept.41 If this makes sense, then in some cases, a reformed concept might provide a better way of thinking and talking about that very topic. For example: one might think that mass concepts informed by general relativity are better ways of thinking about the very same topic (namely, the topic of mass) that we thought about using Newtonian mass concepts. One might think that endorsing this claim is the best way to make sense of a cluster of issues about continuity of inquiry, the presence of certain kinds of substantive disagreement, and various forms of belief report. In light of this, we think that it is at least possible that reformed ethical concepts can potentially provide improved ways of thinking and talking about the same topic—namely, “ethical reality”—that we previously thought and talked about using our existing ethical concepts.

3.3 Further illuminating the range of after metaethics projects

The After Metaethics model we have introduced involves a number of different moving parts. In this subsection, we discuss two of them. Doing so, we argue, can help us get a better sense of what sorts of possibilities there are for engaging in “after metaethics” projects, as well as some of the general challenges facing such projects.

First, consider the kinds of conceptual ethics positions that one might argue for, when one is engaged in an “after metaethics” project (in the bottom right corner of the diagram). For simplicity, in the previous section, we focused on conceptual ethics arguments for “revolutionary” views, which seek to reform or replace our ethical thought and talk. However, as we emphasized in §1, this does not exhaust the defensible responses to finding that a fragment of thought and talk fails to meet salient desiderata. Rather, we might instead argue for retention or abandonment in light of this sort of argument.

Even if we just focus on “revolutionary” views one might argue for, there is a wide range of possible views here. One especially striking possibility is to argue for revolutionary error theory. Here is one possible motivation for considering such a view. Suppose that our ethical thought and talk regularly ends up failing to get certain desiderata we want, but that this isn’t because of the literal (semantic) content of words or concepts. Instead, suppose this is because of relatively robust patterns in the pragmatics of that thought and talk. For example: perhaps, across a wide range of contexts, it regularly leads us to a presupposition or reference failure (of the kind that the canonical error theorist is concerned with), or it regularly leads us to false normative views. Suppose further that the fact that these patterns are in the pragmatics makes it harder to successfully campaign to reform (or replace) the relevant parts of ethical thought and talk. This might be because people (effectively) exploit various linguistic mechanisms to distance themselves from the pragmatically communicated information. In this case, one might seek to engineer a revolutionary error theory, in the hopes that enshrining the relevant errors into the semantic content would make it easier to then successfully campaign for subsequent abandonment or reform of this fragment of thought and talk.

Second, consider the desiderata that might be used in evaluating some ethical thought and talk. Canonical error theorists are characteristically concerned with avoiding reference failure or systematic falsity. The three case studies in the last section illustrate the idea that we might evaluate our ethical thought and talk using a variety of other desiderata. And in the broader literature on conceptual ethics and conceptual engineering, philosophers appeal to a wide range of (purported) desiderata on fragments of thought and talk. These range from “metaphysical” desiderata (such as “carving nature at its joints”) to more “epistemological” ones (such as helping foster successful inquiry) to more “moral” or “political” ones (such as helping in emancipatory projects, or in the pursuit of a more just society).42 Work in the conceptual ethics of ethics might, in theory, draw on any of these sorts of desiderata.

To briefly illustrate another of the possibilities here, consider the epistemological dimension of J. L. Mackie’s so-called “argument from queerness.”43 Mackie argues that if there were “objective” values in his sense of “objective,” then the epistemology required to know about them would be “utterly different from our ordinary ways of knowing everything else.”44 Suppose that the “metaphysical” part of Mackie’s argument fails, and that there really are “objective” values in Mackie’s sense. His epistemological argument might then suggest that we have no way of knowing what the “objective” value facts are. This sort of conclusion is natural grist for a conceptual ethics argument. For it might seem plausible that it is counterproductive to use ethical thought and talk that picks out epistemically inaccessible properties. And one could argue that we should reform this thought and talk to avoid this problem.

4. Authoritative Normativity and Metanormative Error Theory

So far in this paper, we have argued that we can usefully generalize from the structure of “after error” arguments in metaethics to develop an illuminating After Metaethics model for thinking about how metaethics and the conceptual ethics of ethics interact. This section explores a way that a broader metanormative error theory can matter for conceptual ethics projects.

As we have noted, conceptual ethics involves using norms (or values, etc.) in evaluating conceptual ethics proposals. For example, one could use moral norms in evaluating such proposals, or instead use the rules of a child’s treehouse club. As these examples illustrate, the question of which norms to use in evaluating conceptual ethics projects is a crucial question for conceptual ethics.45

One way of answering this question takes its cue from other areas of philosophy that involve significant amounts of normative inquiry, including ethics, epistemology, aesthetics, and political philosophy. For many philosophers, part of what makes these areas important is that they seem to (at least often) involve investigating and deploying normative standards that are “authoritative”—ones that characterize what “really matters.”46 This sort of authoritative normativity seems to contrast with the “normativity” of a given inquirer’s idiosyncratic standard, or the treehouse club norms just mentioned. If conceptual ethics is to be a serious form of normative inquiry, then it might seem that the norms deployed in conceptual ethics should either themselves be authoritative, or be such that their use is endorsed by an authoritative norm.

This idea interacts with our discussion in §3.2 of how well-known metaethical views might be advanced as revolutionary views in conceptual ethics. One motive for engaging in conceptual ethics might be the thought that the norms picked out by some alternative to our current ethical concepts (e.g.) might be more authoritative than the norms picked out by the latter concepts. And this might affect which norms one uses in arguing for the revolutionary view. For example, consider the sort of “revolutionary Kantian constructivism” we mentioned above. Suppose that one thought that the norm picked out by the new normative concept one is arguing for is more authoritative than the ethical norms picked out by our current concepts. Then one might want to argue on behalf of this view using the norms picked out by the new normative concept one is arguing for.

The idea that conceptual ethics projects should use authoritative norms, however, crucially presupposes that there are authoritative norms. This idea is challenged by nihilism about authoritative normativity. Such nihilism is an important component of some metanormative error theories. For example, Bart Streumer has recently argued for an error theory that extends to all normative thought and talk that has (put intuitively) the appearance of tracking anything normatively “important,” including instrumental and prudential normativity.47 Given the motivations we sketched above for appealing to authoritative normativity when doing conceptual ethics, such a nihilism seems to threaten to undercut the normative foundations of conceptual ethics as a whole.48

Someone enthusiastic about conceptual ethics projects might think that the solution is simply to introduce a new normative concept that does pick out an authoritative norm. But nihilism about authoritative normativity is a metaphysical claim. If there is nothing authoritative to pick out, then the attempted introduction will fail.49 (Compare: ‘phlogiston’ was introduced as a theoretical term, but this introduction failed, because there simply was nothing that matched the relevant theoretical role.)

Consider two salient ways forward if nihilism about authoritative normativity were true. First, as in the specifically ethical case, one might defend retentionism in the face of metanormative error theory: perhaps it makes sense (in some way) to continue to use authoritatively normative concepts generally (or just in conceptual ethics) despite the error theory.50 It is hard to see how this is attractive: can it really be appealing for the project of conceptual evaluation and revision to proceed on the basis of crucial false beliefs or presuppositions?

Second, one might simply grant that we cannot meet the motivation introduced in this section. On this picture, conceptual ethics projects can be evaluated relative to any of a wide variety of norms, none of which are authoritative.51 There are various interesting descriptive facts we can cite about such standards (e.g., psychological or sociological facts about people’s relations to them) but no fundamental, non-perspectival normative asymmetries between them. The best that might be said for such a norm might be that the norm endorses its own use as a norm for conceptual ethics projects.52 It strikes us as disappointing if this sort of “self-endorsement” is the best that we can do by way of advocating for the use of a norm in the context of doing conceptual ethics.53

Some working in conceptual ethics might be fine with either of these options, and might well have endorsed one of them all along. We ourselves take both of them to be relatively bleak possibilities that we hope we are not forced to accept. In other work, we hope to be able to show why we are not forced to this sort of perspective on conceptual ethics. But we do not aim to adjudicate among these options here. Rather, our aim in this section has been to motivate a view about which kinds of norms (or values, etc.) one should use when doing conceptual ethics, and then, in turn, illustrate another important (and relatively unexplored) way in which metanormative inquiry and conceptual ethics inquiry can interact.

Conclusion

In this paper, we’ve explored some of the relations between metaethical error theory and the conceptual ethics of normativity. Our guiding thought has been that discussions around metaethical error theory—and in particular, discussion of the “after error” question—can help illuminate some of the general ways that metanormative inquiry and the conceptual ethics of normativity can interact.

Sustained, explicit reflection on the conceptual ethics of normativity is a relatively young enterprise. With that in mind, we want to stress a point we made earlier: namely, that we by no means take our discussion here of the ways these projects can interact to be exhaustive. Indeed, we take there to be a range of further ways they can interact, including some that we explore at more length in other work. We hope our work in this paper helps to spur further investigation into these myriad connections.

In closing, we want to flag the following point about the potential broader philosophical relevance of our work in this paper. Our focus has been squarely on issues in ethical theory. One important question this paper prompts is whether and to what extent the After Metaethics model we present here might be adapted to illuminate discussions in other subareas of philosophy, such as epistemology or metaphysics. We think there is rich terrain here to explore, especially given what we think is the often-neglected importance of “conceptual ethics” arguments in many areas of philosophy.54 Our hope is that our arguments here can help spur further reflection on the (actual or potential) role of arguments in conceptual ethics in other areas of philosophy.

Notes

  1. This phrase draws from (McPherson and Plunkett 2020). For an overview of recent work on conceptual ethics (and the closely connected topic of “conceptual engineering”), see the essays collected in (Burgess, Cappelen, and Plunkett 2020). ⮭
  2. (Railton 1986). ⮭
  3. We use the term ‘canonical error theory’ because, as we illustrate later in the paper, there are many potential ways that a fragment of thought and talk might in some sense enshrine an “error.” The “canonical” form of the view we introduce in the text is only one influential and salient possibility. Contemporary canonical error theorists include (Olson 2014) and (Streumer 2017). (Mackie 1977) is often represented as the paradigmatic canonical error theorist. ⮭
  4. For a helpful gloss on error theory, which overlaps with our characterization of canonical error theory, see (Olson 2017). ⮭
  5. For these alleged implications of our ethical concepts, see, respectively, (Mackie 1977), (Olson 2014), and (Joyce 2001). This is not the only way to argue for error theory. For example, prominently, (Streumer 2017) offers an importantly different style of argument. ⮭
  6. See (Lutz 2014) for helpful recent discussion of this question. ⮭
  7. For a more extensive discussion of the recent responses to this issue, see (Jaquet 2020). For further discussion of a range of these options, see the essays collected in (Garner and Joyce 2018). ⮭
  8. See (Burgess and Plunkett 2013b) for the suggestion to use the term ‘eliminativism’ for the idea that we should abandon using a certain kind of thought and talk, at least in certain contexts, for certain purposes. See (Cappelen 2023) for the use of ‘abandonment’ to refer to this idea. ⮭
  9. See (Garner 2007). One might instead argue for abandonment on the basis that ethical thought and talk generally has bad effects. This idea is defended by (Hinckfuss 1987). ⮭
  10. See (Olson 2014) and (Streumer 2017). ⮭
  11. For general discussion about retention in some but not all contexts, see (Burgess and Plunkett 2013b) and (Cappelen and Plunkett 2020). ⮭
  12. See (Joyce 2001) for revolutionary fictionalism, and (Köhler and Ridge 2013) and (Svoboda 2017) for revolutionary expressivism. ⮭
  13. In this paper, we use single quotation marks (e.g., ‘cat’) strictly to mention linguistic items. As we have already been doing throughout the paper, we use double quotation marks (e.g., “cat”) for a variety of tasks including quoting others’ words, scare quotes, and mixes of use and mention. ⮭
  14. For a much more careful explanation of this account, see (McPherson and Plunkett 2017), where our focus is ultimately on metaethics in particular. See also (Plunkett and Shapiro 2017), (McPherson and Plunkett 2021a), and (McPherson and Plunkett 2024a). ⮭
  15. See (Burgess and Plunkett 2013a), (Burgess and Plunkett 2013b), and (Burgess and Plunkett 2020). See also (McPherson and Plunkett 2020), (McPherson and Plunkett 2021a), and (McPherson and Plunkett 2024a). A closely connected topic is “conceptual engineering.” Put roughly, we take projects in “conceptual engineering” to combine work in conceptual ethics with work on “conceptual innovation” (e.g., designing conceptual changes, or new concepts) and “conceptual implementation” (e.g., trying to actually get people to use reformed concepts). See (Cappelen and Plunkett 2020) and (Burgess and Plunkett 2020) for accounts of conceptual engineering along these lines. ⮭
  16. Here we follow (Burgess and Plunkett 2013a), who, when introducing the label ‘conceptual ethics’, emphasize its potential misleadingness, but nonetheless hold that it is the best label for their purposes. ⮭
  17. See (McPherson and Plunkett 2024a). For connected discussion, see (Eklund 2017). ⮭
  18. This kind of observation is part of why, in (McPherson and Plunkett 2017), we favor a context-sensitive account of what counts as a metaethical “claim” (or “theory,” “question,” etc.). Our idea, in short, is that, in different contexts, different relations to the overall metaethical project will be salient, and be the basis for determining what counts as a metaethical “claim” etc. This idea can naturally be extended to what counts as “conceptual ethical claim” (or “theory,” “question,” etc.). ⮭
  19. For examples of other options, see (McPherson and Plunkett 2024a, §§5-6) and (McPherson and Plunkett 2021a, §7). ⮭
  20. As demonstrated by the basic overview of what metaethical error theory involves in (Olson 2017). ⮭
  21. For example, see (Ingram 2015). ⮭
  22. For example, this thought is arguably part of what motivates “quasi-realist” versions of expressivism, such as in (Blackburn 1998) and (Gibbard 2003). ⮭
  23. For discussion of other prominent arguments that are often thought of as “metaethical,” but that might well be also (and perhaps better) developed as ones primarily in conceptual ethics, see (McPherson and Plunkett 2024a). See also (Eklund 2017) for connected discussion (and for general reflection on what we are here calling the “conceptual ethics of normativity”). ⮭
  24. (Korsgaard 1996). ⮭
  25. The term ‘realism’ is used in a variety of cross-cutting ways in metaethics. For discussion, see (Dunaway 2017). ⮭
  26. See (Korsgaard 1996, 9-13). For connected discussion, see (McPherson 2020), which argues that a challenge very closely connected to Korsgaard’s—the challenge of securing the “deliberative authority” of our normative thought—is one half of the deepest challenge to normative realism. ⮭
  27. (Korsgaard 1996, 34). ⮭
  28. (Enoch 2011a, 209). ⮭
  29. (Enoch 2006). ⮭
  30. (Enoch 2011b). ⮭
  31. For illustrative discussion of this and other standard objections to subjectivism, see (McPherson and Plunkett 2017). ⮭
  32. It should be noted that one sort of bridge premise that might enable us to (fairly directly) draw a metaethical conclusion about our actual ethical concepts (in both this case and the other two we discuss in this section) is the metaethical premise that our actual ethical concepts are the concepts that it is best for us to use. For defense of related ideas, see (Enoch 2009), (Preston-Roedder 2014), and (Sayre-McCord Manuscript). For further discussion of such views (as well as other ones that claim normative facts playing a role in determining the content of our actual normative concepts), and how such views interact with the distinction between metaethics and the conceptual ethics of normativity that we are working with in this paper, see (McPherson and Plunkett 2024a). ⮭
  33. For a helpful overview of “constructivism” in metaethics, see (Barry 2017). ⮭
  34. For doubts about whether Korsgaard herself has offered the relevant sorts of arguments, see (Hussain and Shah 2006). ⮭
  35. See (McPherson and Plunkett 2024a) for complementary discussion of how certain “constructivist” ideas from Korsgaard and from Street (Street 2006) can be interpreted as claims in conceptual ethics. ⮭
  36. It is relatively common in the literature on moral fictionalism to distinguish “revolutionary” from “hermeneutic” fictionalism, drawing on (Burgess 1983). (See, for example, (Kalderon 2005)). Our point is, again, that philosophical tools most salient in the “after error” context (e.g., discussions of “revolutionary” forms of views such as fictionalism, expressivism, etc.) are of quite general significance to work on the foundations of ethics. ⮭
  37. This sort of possibility is explored by (Kahane 2013). ⮭
  38. Parfit uses this point not to characterize this as an objection to the relevant metanormative views. Rather, he uses it to dramatize what is at stake in certain central metanormative debates. See (Parfit 2011, Vol. 2, 303-304). For a critical response, see (Schroeder 2017). ⮭
  39. For an introduction to debates about how different metaethical theories can and cannot explain disagreement, see (Björnsson 2017). ⮭
  40. We here draw on (McPherson and Plunkett 2017). ⮭
  41. See, e.g., (McPherson and Plunkett 2021c) and (Cappelen 2018). ⮭
  42. For an example of appeal to broadly “metaphysical” norms or values when doing (what we take to be) conceptual ethics, see (Sider 2011); for appeal to broadly “epistemic” ones, see (Scharp 2013); and for broadly “moral” or “political” ones, see (Haslanger 2000). For further discussion of the range of values and norms that animate discussion in conceptual ethics, see (Burgess and Plunkett 2013b) and (Cappelen and Plunkett 2020). ⮭
  43. (Mackie 1977). ⮭
  44. (Mackie 1977, 38). ⮭
  45. We discuss this topic at length in (McPherson and Plunkett 2024b). ⮭
  46. For further discussion of the idea of an authoritative norm, see (McPherson 2018) and (McPherson and Plunkett Forthcoming). ⮭
  47. (Streumer 2017, §50 and §52). ⮭
  48. For connected reflections, see (McPherson and Plunkett 2021b). ⮭
  49. Here, for simplicity, we are simply assuming that any authoritatively normative concept will be representational. ⮭
  50. It is worth noting another kind of option here for retentionism. Streumer (2017) argues for retaining our normative concepts based on his striking idea that, even though metanormative error theory is true, we can’t believe it. ⮭
  51. For the related general idea of “deflationary normative pluralism,” see (Tiffany 2007). ⮭
  52. See also connected discussion in (Burgess 2020), drawing on earlier discussions in (Burgess and Plunkett 2013b), about connected issues of “self-abnegating” concepts, rather than “self-vindicating” ones. ⮭
  53. Issues about the limits of the kind of “self-vindication” we just discussed might in fact pose deeper worries about reliance on our “authoritatively normative” thought and talk as well, and whether it is really getting at what we might (put intuitively) think of as the “normatively important” parts of reality. For further discussion, see (McPherson and Plunkett 2021b) and (Eklund 2017). See also connected discussion in (Burgess 2020), drawing on earlier discussions in (Burgess and Plunkett 2013b) about connected issues of “self-abnegating” concepts, rather than “self-vindicating” ones. ⮭
  54. For a recent exploration of the role of conceptual ethics in epistemology, see (McPherson and Plunkett 2021a), and for a recent exploration of the role of conceptual ethics in metaphysics, see (Thomasson 2017). ⮭

Acknowledgments

Thanks to Tom Adams, Sonu Bedi, Mitch Berman, Alexis Burgess, Herman Cappelen, Nico Cornell, Natalie Dokken, Andy Egan, David Enoch, Jesse Ferraioli, Chris Howard, Zoë Johnson King, Niko Kolodny, Anders Knospe, Zachary Lang, Stephanie Leary, Hille Paakkunainen, Philip Pettit, Matthieu Queloz, Toni Rønnow-Rasmussen, Timothy Rosenkoetter, Adrian Russian, Thomas Schmidt, Zoe Thierfelder, Amie Thomasson, Patrick Todd, Manuel Vargas, Mark Van Roojen, Daniel Wodak, and anonymous referees for helpful feedback and discussion. Thanks also to participants in discussions about previous versions of this paper at Bielefeld University, University of Bucharest, University of California, San Diego, Dartmouth College, Humboldt University of Berlin, University of Lisbon, University of Lund, Montreal Metaethics Workshop, University of Oxford, Philosophy Desert Workshop, the Swiss Philosophical Society, and Syracuse University. Thanks also to the Wissenschaftskolleg zu Berlin for fellowship funding for one of us (Plunkett), which helped support our work on this article.

Works Cited

Barry, Melissa. 2017. Constructivism. In The Routledge Handbook of Metaethics, edited by T. McPherson and D. Plunkett. New York: Routledge. 385–401.

Björnsson, Gunnar. 2017. The Significance of Ethical Disagreement for Theories of Ethical Thought and Talk. In The Routledge Handbook of Metaethics, edited by T. McPherson and D. Plunkett. New York: Routledge. 275–291.

Blackburn, Simon. 1998. Ruling Passions. Oxford: Clarendon Press.

Burgess, Alexis. 2020. Never Say ‘Never Say “Never”’? In Conceptual Engineering and Conceptual Ethics, edited by A. Burgess, H. Cappelen and D. Plunkett. Oxford: Oxford University Press. 125–131.

Burgess, Alexis, Herman Cappelen and David Plunkett. 2020. Conceptual Engineering and Conceptual Ethics. Oxford: Oxford University Press.

Burgess, Alexis, and David Plunkett. 2013a. Conceptual Ethics I. Philosophy Compass 8 (12): 1091–1101.

Burgess, Alexis, and David Plunkett. 2013b. Conceptual Ethics II. Philosophy Compass 8 (12): 1102–1110.

Burgess, Alexis, and David Plunkett. 2020. On the Relation Between Conceptual Engineering and Conceptual Ethics. Ratio 33 (4): 281–294.

Burgess, John P. 1983. Why I am Not a Nominalist. Notre Dame Journal of Formal Logic 24 (1): 93–105.

Cappelen, Herman. 2018. Fixing Language: An Essay on Conceptual Engineering. Oxford: Oxford University Press.

Cappelen, Herman. 2023. The Concept of Democracy: An Essay on Conceptual Amelioration and Abandonment. Oxford: Oxford University Press.

Cappelen, Herman and David Plunkett. 2020. A Guided Tour of Conceptual Engineering and Conceptual Ethics. In Conceptual Engineering and Conceptual Ethics, edited by A. Burgess, H. Cappelen and D. Plunkett. Oxford: Oxford University Press. 1–26.

Dunaway, Billy. 2017. Realism and Objectivity. In The Routledge Handbook of Metaethics, edited by T. McPherson and D. Plunkett. New York: Routledge. 135–150.

Eklund, Matti. 2017. Choosing Normative Concepts. Oxford: Oxford University Press.

Enoch, David. 2006. Agency, Shmagency: Why Normativity Won’ t Come from What Is Constitutive of Action. Philosophical Review 115 (2): 169–198.

Enoch, David. 2009. Wouldn’ t it be nice if P, therefore, P (for a moral P). Utilitas 21 (2): 222–224.

Enoch, David. 2011a. Shmagency Revisited. In New Waves in Metaethics, edited by M. Brady. London: Palgrave Macmillan. 208–233.

Enoch, David. 2011b. Taking Morality Seriously: A Defense of Robust Realism. Oxford: Oxford University Press.

Garner, Richard. 2007. Abolishing Morality. Ethical Theory and Moral Practice 10 (5): 499–513.

Garner, Richard and Richard Joyce. 2018. The End of Morality: Taking Moral Abolitionism Seriously. New York: Routledge.

Gibbard, Allan. 2003. Thinking How to Live. Cambridge, MA: Harvard University Press.

Haslanger, Sally. 2000. Gender and Race: (What) Are They? (What) Do We Want Them to Be? Noûs 34 (1): 31–55.

Hinckfuss, Ian. 1987. The Moral Society: Its Structure and Effects. Canberra: Australian National University.

Hussain, Nadeem J. Z., and Nishi Shah. 2006. Misunderstanding Metaethics: Korsgaard’ s Rejection of Realism. In Oxford Studies in Metaethics, Volume 1, edited by R. Shafer-Landau. New York: Oxford University Press. 265–294.

Ingram, Stephen. 2015. After Moral Error Theory, After Moral Realism. Southern Journal of Philosophy 53 (2): 227–248.

Jaquet, François. 2020. Sorting Out Solutions to the Now-What Problem. Journal of Ethics and Social Philosophy 17 (3): 239–258.

Joyce, Richard. 2001. The Myth of Morality. Cambridge: Cambridge University Press.

Kahane, Guy. 2013. Must Metaethical Realism Make a Semantic Claim? Journal of Moral Philosophy 10 (2): 148–178.

Kalderon, Mark Eli. 2005. Moral Fictionalism. Oxford: Clarendon Press.

Köhler, Sebastian and Michael Ridge. 2013. Revolutionary Expressivism. Ratio 26 (4): 428–449.

Korsgaard, Christine M. 1996. The Sources of Normativity. New York: Cambridge University Press.

Lutz, Matt. 2014. The ’ Now What’ Problem for error theory. Philosophical Studies 171 (2): 351–371.

Mackie, J. L. 1977. Ethics: Inventing Right and Wrong. New York: Penguin.

McPherson, Tristram. 2018. Authoritatively Normative Concepts. In Oxford Studies in Metaethics Vol. 13, edited by R. Shafer-Landau. Oxford: Oxford University Press. 253–277.

McPherson, Tristram. 2020. Deliberative Authority and Representational Determinacy: A Challenge for the Normative Realist. Ergo 6 (45): 1331–1358.

McPherson, Tristram and David Plunkett. 2017. The Nature and Explanatory Ambitions of Metaethics. In The Routledge Handbook of Metaethics, edited by T. McPherson and D. Plunkett. New York: Routledge. 1–25.

McPherson, Tristram and David Plunkett. 2020. Conceptual Ethics and The Methodology of Normative Inquiry. In Conceptual Engineering and Conceptual Ethics, edited by A. Burgess, H. Cappelen and D. Plunkett. Oxford: Oxford University Press. 274–303.

McPherson, Tristram and David Plunkett. 2021a. Conceptual Ethics, Metaepistemology, and Normative Epistemology. Inquiry.

McPherson, Tristram and David Plunkett. 2021b. Evaluation Turned on Itself: The Vindicatory Circularity Challenge to the Conceptual Ethics of Normativity. In Oxford Studies in Metaethics, Vol. 16, edited by R. Shafer-Landau. Oxford: Oxford University Press. 207–232.

McPherson, Tristram and David Plunkett. 2021c. Topic Continuity in Conceptual Engineering and Beyond. Inquiry.

McPherson, Tristram and David Plunkett. 2024a. Metaethics and the Conceptual Ethics of Normativity. Inquiry 67 (1): 93–126.

McPherson, Tristram and David Plunkett. 2024b. Normative Standards and the Epistemology of Conceptual Ethics. Inquiry 67 (3): 954–984.

McPherson, Tristram and David Plunkett. Forthcoming. Authoritative Normativity. In The Oxford Handbook of Meta-ethics, edited by D. Copp and C. S. Rosati.

Olson, Jonas. 2014. Moral Error Theory: History, Critique, Defence. Oxford: Oxford University Press.

Olson, Jonas. 2017. Error Theory in Metaethics. In The Routledge Handbook of Metaethics, edited by T. McPherson and D. Plunkett. New York: Routledge. 58–71.

Parfit, Derek. 2011. On What Matters. Oxford: Oxford University Press.

Plunkett, David and Scott Shapiro. 2017. Law, Morality, and Everything Else: General Jurisprudence as a Branch of Metanormative Inquiry. Ethics 128 (1): 37–68.

Preston-Roedder, Ryan. 2014. A Better World. Philosophical Studies 168 (3): 629–644.

Railton, Peter. 1986. Moral Realism. The Philosophical Review 95: 163–207.

Sayre-McCord, Geoff. Manuscript. On a Theory of a Better Morality.

Scharp, Kevin. 2013. Replacing Truth. Oxford: Oxford University Press.

Schroeder, Mark. 2017. What Matters About Meta-Ethics? In Does Anything Really Matter?: Essays on Parfit on Objectivity, edited by P. Singer. Oxford: Oxford University Press. 213–230.

Sider, Theodore. 2011. Writing the Book of the World. Oxford: Oxford University Press.

Street, Sharon. 2006. A Darwinian Dilemma for Realist Theories of Value. Philosophical Studies 127 (1): 109–166.

Streumer, Bart. 2017. Unbelievable Errors: An Error Theory About All Normative Judgments. New York: Oxford University Press.

Svoboda, Toby. 2017. Why Moral Error Theorists Should Become Revisionary Moral Expressivists. Journal of Moral Philosophy 14 (1): 48–72.

Thomasson, Amie L. 2017. What Can We Do, When We Do Metaphysics? In The Cambridge Companion to Philosophical Methodology, edited by G. d’ Oro and S. Overgaard. Cambridge: Cambridge University Press. 101–121

Tiffany, Evan. 2007. Deflationary Normative Pluralism. Canadian Journal of Philosophy 37 (5): 231–262.