Skip to navigation – Site map

HomeIssuesXIII-2Symposium. Pragmatist Ethics: The...Pragmatism and Moral Objectivity

Symposium. Pragmatist Ethics: Theory and Practice

Pragmatism and Moral Objectivity

Michael Klenk


Most non-robust-realist metaethical theories, such as expressivism, constructivism, and non-robust forms of realism, claim to retain a sense of objectivity in ethics. A persistent issue for these theories is to identify an objective criterion for moral truth that meets their objectivist aspiration.

Objectivist aspirations are often probed by confronting non-realists with abject normative positions, such as those of rational racists, which are licensed by the framework of the respective non-realist theory but nevertheless strike us a wrong. In such cases, non-realist theories face a dilemma. Either they allow that anything goes and thereby forgo their objectivist aspirations or they disallow abject normative positions. In the latter case, however, they have nothing to turn to but subjective criteria ultimately related to one’s personal outlook. This is unacceptably smug.

I argue that pragmatism in the spirit of Charles S. Peirce partially solves this dilemma. True belief would withstand experience and argument were we to inquire as far as we fruitfully could on the matter. I elucidate this notion and argue that pragmatist construal of moral truth provides a substantive, objective criterion to determine the truth value of moral claims, without recourse to subjective criteria. This puts pragmatism ahead of rival non-realist theories.

Top of page

Author's notes

I am grateful to two anonymous reviewers and Sarin Marchetti for helpful comments on an earlier version of this paper.

Full text

1. Introduction

1The case of objectivity in ethics is a vexed issue – making sense of it resembles a horse trade: get objective moral standards, but pay with your naturalistic conscience. Those who bagged moral objectivity have paid with a problematic metaphysics and epistemology (robust moral realism, e.g. Enoch 2011, Wielenberg 2014, Shafer-Landau 2003), incurred the commitment of believing the necessity of valuing human nature (Kantian constructivism, e.g. Korsgaard 1998), or left indebted with the task to defend notions of agreement (procedural constructivism, e.g. Rawls 1980, Habermas 1991) and conceptions of idealised states of consideration (e.g. Smith, Lewis & Johnston 1989, Railton 1986). Those who left the market for moral objectivity empty handed will either go for some third-party product that promises to do the same with less (advanced versions of non-cognitivism, Gibbard 2003, Blackburn 1998) or else leave sobered and disenchanted: they’ll lament the hype, renounce it, and perhaps try to explain just why everyone is so excited about objectivity in ethics, while there really is nothing to be sought (Humean constructivism like Street 2008, perhaps error theory, e.g. Joyce 2001). No doubt, many of those who think they got the real thing will be happy with their purchase. But who would not be in for a little bargain?

2Pragmatists offer just that. They tell you what you really want and then show you how you can have it. In the market for moral objectivity, pragmatists are clever marketers and therapists at the same time – they claim to listen to our needs, free us from faulty presuppositions, and then deliver what fits our cleansed outlook.

  • 1 The pragmatism that I’ll discuss here is thus in stark contrast to Rorty’s nihilist view, which is (...)

3My aim in this paper is to show that a version of pragmatism inspired by Charles S. Peirce makes sense of objectivity in ethics, without one of the costs incurred by rival theories. I will discuss the notion of objectivity as developed by Cheryl Misak, who has defended Peircean pragmatism in metaethics (Misak 2002). Peircean pragmatism relies on a reinterpretation of the notion of “truth” and thereby allows us to speak sensibly of truth when rival accounts have long given up.1

  • 2 However, see Wilson 2010 for a recent attempt.

4The distinctive approach of pragmatism is not considered appropriately in recent metaethics. A widely cited, and still relevant, overview of the metaethical debate does not mention it (Darwall, Gibbard & Railton 1992), and neither does the now standard introduction to metaethics (Miller 2007).2 The impasse of metaethics in regards to the question of moral objectivity is, however, so widely noted and accepted that it is merited to explore this new approach in greater detail.

5I assume in this essay that robust moral realism is untenable – it might be rejected for either of two reasons: metaphysical (in the spirit of Mackie 1977) or epistemological (in the spirit of Benacerraf 1973; Field 1980, see also Klenk 2022). The latter has recently been applied in the metaethical case (Street 2006) under the hood of evolutionary debunking arguments. When I speak of realism, I mean robust moral realism (which entails that there are mind-independent moral properties and facts). When I speak of non-realist theories, I mean all theories that are not robust realist theories.

1.1. Problem Statement

6All metaethical theories, unless they are committed to moral nihilism, have to explain why some moral positions are better than others. The trouble for non-realist theories of morality is how to account for this demand. A preliminary concern is the ambiguity of the challenge. Timmons, who dubbed the worry about subjectivism in respect to non-realist theories the anything goes challenge (AGC), demands that “we should ask ourselves just what counts as a philosophically respectable reply to the challenge” (Timmons 1991: 393). My first aim is to do just this by bringing the AGC into perspective.

  • 3 The challenge relates to the challenge of rational egoism in first-order normative ethics: Thrasyma (...)

7On the one hand, the AGC cannot be construed so that only a reference to robust realist moral facts and properties can solve it. This would beg the question against non-realist theories. On the other hand, the challenge’s solution cannot be an account of disagreement, for that is possible even for the simplest subjectivist theories. The challenge is commonly advanced by imagining characters that are at odds with our first-order normative beliefs about good conduct, such as rational Nazis or an ideally coherent Caligula.3 As you will see, there are strange and repugnant characters in store for all non-realist theories.

8The core problem that I address is that pragmatism might have trouble answering the AGC, too. If pragmatism fails to give an answer to the AGC, then it fares no better than rival theories and, insofar as a solution to the AGC concerned, there is no reason to adopt pragmatism rather than one of its rival theories. If pragmatism does better in answering the AGC, however, then this gives pragmatism important leverage because it solves a pervasive and relevant metaethical problem.

1.2. Argument and Structure

9I use a bit of an idiosyncratic terminology in this essay. Victim Theories (VT) are metaethical theories that meet two criteria. They fail the anything goes challenge (AGC) and they are committed to retaining a sense of objectivity in ethics. I understand “objectivity in ethics” to rely on the following two claims:

  • 4 By “ultimately” I mean that the final answer to repeated iterations of “why” questions about moral (...)

Evaluation: some moral claims are wrong, and others are right, and the relevant assessment does not ultimately4 depend on ideosyncratic outlooks.

  • 5 They might also be “on par,” see Chang 2002. In any case, the relevant criteria (metaphysical, conc (...)

Comparability: moral states of affairs, actions, and characters can be compared: they are better than, worse than, or equally good as others, and the relevant assessment does not ultimately depend on ideosyncratic outlooks.5

10Whatever moral claim is not objective in the above sense is subjective. I argue as follows:

  • 6 A third option is that VTs abandon the project of accounting for moral truth in some sense. The res (...)

1. Victim Theories are subjectivist (anything goes) or smug about the criterion for moral truth and falsity (smugness).6

2. Neither subjectivism nor smug meta-subjectivism can be squared with the objectivist aspirations of Victim Theories.

3. Peircean pragmatism is neither subjectivist nor smug.

4. So, Peircean pragmatism is not a Victim Theory.

  • 7 In between realist objectivity and subjectivism.

11Before going through the argument, I introduce the core tenants of Cheryl Misak’s account of Peircean pragmatism in section 2. In section 3, I demonstrate the pervasiveness and relevance of the AGC by reviewing objections to four different kinds of non-realist metaethical theories (sophisticated expressivism, constructivism, reductive realism, and internal realism). Although all VTs differ in their specific approach to moral objectivity, they fall prey to the same worry: VTs are subjectivist although they aspire to be objectivist. Section 4 establishes the second horn of my first premise: VTs may avoid the charge that anything goes, but their only means of establishing this is smug because one has to rely on one’s personal outlook to establish that oneself is right and others likely to be mistaken. In section 5, I turn to Peircean pragmatism and its version of an objective criterion for moral truth and falsity. I argue that it does indeed constitute an objective criterion in the sense demanded in this essay. It offers a genuine in-between notion of objectivity7 because the criterion for what is objectively right or wrong is never dependent on one’s point of view. Rather, the criterion relies on a Peircean account of truth as what would be “indefeasible by experience.” I consider objections in section 6. Finally, I conclude that pragmatism is not a Victim Theory. It makes sense of genuine moral objectivity and needs not be smug to do so. The victory is limited, however. Pragmatism delivers an objective criterion of moral truth, and in that sense exceeds competing non-realist theories, but worries remain regarding several aspects of this notion.

2. Peircean Pragmatism

  • 8 Henceforth I use pragmatism to denote the pragmatism championed by Misak in the spirit of Peirce. T (...)

12Cheryl Misak defends a version of pragmatism inspired by Charles S. Peirce. Misak clarifies, solidifies, and enhances Peirce’s pragmatism in several ways, but the essential commitments of pragmatism remain the same. First, a reconstruction of the notion of moral truth that makes moral truths knowable a posteriori and essentially linked to human experience. Second, an analysis of the nature of assertion and the claim that assertion is tied to truth. Third, a rejection of robust moral realism on epistemological grounds and, fourth, the commitment to retain a sense of objectivity and the ability to speak of truth and falsity in ethics. The latter two commitments make pragmatism relevant for my project and, at the same time, threaten it with being just another Victim Theory. The former two, however, enable pragmatism to escape this charge, or so I argue.8

13In Misak’s words, her account is an attempt to “make sense of our standards of rationality, truth, and value as genuinely normative and binding while recognising that they are profoundly human phenomena” (Misak 2013: xi). Misak’s account rests on three core claims that exploit the links between assertion, belief, and truth (compare Heney 2016, who argues in similar vein but on different grounds that our practices of moral deliberation are driven by cognitivist aspirations). Misak´s first argument aims to establish the nature of assertion. It takes the form of a transcendental argument and goes as follows:

Misak Assertion:

  • 9 Thus, this also applies if one is trying to mislead with a false assertion. Lying, for instance, de (...)

1. We take truth to be our aim when we assert and when we inquire (Misak 2002: 60ff.).9

  • 10 A belief aims at truth iff I believe p, I believe p to be true (Wiggins 2002). If this is right the (...)

2. If one genuinely asserts, then one is aiming at truth.10

3. So, whoever genuinely asserts aims at truth.

  • 11 There might be worries about lying and bullshitting in Frankfurt’s sense and Misak does not address (...)

14Misak’s claim about the nature of assertion is motivated by an analysis of ordinary speech acts and the function of assertion in communication. I will assume its truth for the sake of argument.11 Misak’s second argument is an elaboration of a Peircean account of truth.

Misak Truth:

1. True belief is such that it would be the best belief were we enquire as far as we fruitfully could on the matter (Misak 2013: 37); (Misak 2002: 49).

2. The best belief is such that it would “best fit with all experience and arguments” and “withstand doubt” in this sense (Misak 2002: 49).

3. So, true belief would withstand doubt.

15Henceforth, I shall use “pass the test of experience” as shorthand for the pragmatist doctrine about true belief. Such a belief would not be doubted, it could not be improved upon, and it would best fit with all experience and arguments. If no experience at all could change a belief, then it is not worth investigating it; it would be, as Wittgenstein put it, “a cog upon which nothing turns” (cited in Misak 2002: 51). Misak’s third argument concerns the range of experience relevant to assessing the truth of beliefs.

Misak Experience:

1. Belief, inquiry, and assertion aim at truth.

2. True beliefs are such that they “pass the test of experience,” which entails that they could not be improved upon.

3. Belief, inquiry, and assertion thus aim at beliefs that could not be improved upon.

4. To form beliefs that could not be improved upon, one has to take into account all available evidence.

5. One’s own experience is not all available evidence there is.

  • 12 “Logic inexorably requires that our interests [and experiences] shall not be limited. They must not (...)

6. So, the experience of all experiencers counts.12

16Taken together, Misak’s three arguments yield the following view. All believers aim at truth and the truth is that which would not be overturned by experience. Since truth is tied to experience, and all believers aim at truth, it follows that experience should be understood as wide as possible and encompass the experience of all believers. Passing the test of experience broadly understood would yield the greatest possible resilience of one’s belief, and thus serve the believer’s aim in the best possible way. In our effort to understand reality, “each of us is an insurance company” (CP 5.354) and we must cast our net for being open to experiences and arguments “as wide as the community can turn out to be” (CP 5.357, Misak 2013: 37). Note that the pragmatist does not provide an analytical or nominal definition of truth, since truth is not defined as what will be agreed upon in the long run (Misak 2002: 59).

3. Either Anything Goes According to Victim Theories or…

  • 13 Recall that non-sophisticated expressivism such as that of Ayer (1971 [1936]) has no ambition to re (...)
  • 14 It might be questioned whether the anything goes challenge is worth answering. There are two replie (...)

17Non-realist accounts routinely face objections to the effect that their objectivist appearance is an illusion that makes way for a subjectivist core once probed accordingly. Variants of this challenge are put forward against reductionist realism, sophisticated expressivism,13 constructivism, and conceptual pluralism alike.14 The challenge has real bite only if it frustrates the objectivist aspirations of the respective Victim Theory.

18We will see that Victim Theories can account for moral disagreement in the sense that they explain how it arises and why it irritates us. They may also be able to explain how it is settled on the grounds of first-order moral judgements about the moral error of whomever one has a moral dispute with. But this is not the account of moral disagreement that I am after. Victim Theories cannot, without being smug, evaluate and compare conflicting moral claims.

3.1. Sophisticated Expressivism

19According to sophisticated forms of expressivism, there are no moral facts and moral properties, yet talk of moral truth and falsity is still warranted, even in the case of incompatible normative judgements (e.g. Gibbard 1990, Blackburn 1998). The problems of realism seem avoided, yet the minimal sense of objectivity is retained. Gibbard (2011) provides a neat case-study of such a view. Expressivists, it turns out, do not fare well against the odd and morally repugnant characters that I mentioned in the introduction.

20Gibbard (2011) envisions a case in which a group of people believe that an animal’s pain does not matter for one’s normative and moral considerations (call this position N2). He also envisions that this judgement is entirely coherent with the rest of the beliefs of all individuals of this group as well as with shared knowledge, conventions, and traditions of that group. In contrast, Gibbard himself believes that an animal’s pain does matter (N1). For Gibbard, the features that make an act right or wrong, the features that ground its wrongness, are mind-independent, namely how it feels for a chicken to be in pain. The justification for this thought, however, is agent-dependent. That is, the association of reasons against hurting animals and the fact that they feel pain is dependent ultimately on people’s evaluation of the situation. Expressivists, it seems, meet both the evaluation and the comparability criterion.

21However, expressivists like Gibbard cannot in a relevant sense compare value judgements if the value judgement of the animal tormentors is flawlessly construed according to the expressivist’s own theory. In such cases, expressivists may make sense of disagreement, i.e. they can explain where the disagreement comes from and why it has a troubling feel to it, but they have to accept that the rival position is neither better nor worse than their own position. In that sense, they cannot compare their own value judgements with that of the opposing party and thus violate one of the two demands on objectivity I stipulated above. On Gibbard’s account, he is justified in believing N1 while the animal tormentors are, at the same time, justified in believing N2. However, he is also confident about the animal tormentor’s “wrong” conclusion. Gibbard might, of course, say that the animal tormentors are wrong, according to his standards. But they cannot be wrong according to their own standards. Elsewhere, Gibbard states that the things that make an act wrong are mind-independent, such as the animal’s pain and how it is for an animal to be in pain. But to take these features to constitute reasons in Scanlon’s sense, on which Gibbard relies, is entirely mind-dependent. Gibbard proposes to deal with this scenario by “hypothetical contingency planning” (Gibbard 2011: 40). Roughly, one imagines vividly and with “great perspicacity,” to be one of the animal tormentors and plans to believe N1 instead of N2. Gibbard is not entirely clear about this, but here is an interpretation. The disagreement then consists either in

A) planning (in the actual world) to stick to one’s own normative judgement were one in the shoes of the other,

B) sticking to one’s own normative judgement were one in the shoes of the other.

  • 15 Except, perhaps, one frees oneself from the deterministic course of events and thus achieves Kant’s (...)

22Case B seems impossible to hold on to. If Gibbard really were one of the animal tormentors, would he not have to admit that he’d believe N2? Really being one of the animal tormentors of course entails just taking all their conventions, etc on board as well, and I fail to see how one could arrive at a different moral judgement.15 By hypothesis, the world of the animal tormentors is equal to Gibbard’s world except in that they believe that N2 is correct, while Gibbard believes that N1 is correct, and both are epistemically justified in their respective belief. Gibbard’s analysis then comes down to case A. If two people fundamentally disagree, then, according to Gibbard, they disagree about what they plan to take into account when acting.

23The AGC, it seems, is not about disagreement. Expressivists like Gibbard can account for it by relying on differences in planning. Rather, the problem is that Gibbard has nothing to say to the animal tormentors. They formed their normative position by the rulebook of Gibbard’s expressivism and yet they are doing something that strikes Gibbard, and us I suppose, as wrong. Unfortunately, there is no basis for criticism and Gibbard has to allow that anything goes, as demonstrated by the animal tormentors. On pain of being repetitive: imagine that you live in shared housing. There are rules as to how to behave, e.g. no noise after 10 pm. These rules are equivalent to Gibbard’s expressivist rules. Within the rules, anything goes. Your neighbour stages a rock concert from 4-10pm while you try to watch TV? Well, sadly, you seem to have different conceptions of life, but anything goes (before 10pm) and maybe you’d move out.

3.2. Kantian Constructivism

  • 16 Humean constructivism, as defended by Street 2008, is not a Victim Theory because it does not aspir (...)

24Kantian constructivism16 succumbs to the AGC in similar fashion, although a caveat is in order. Kantian constructivists attempt to secure moral objectivity through an account of universal and categorical reasons that necessarily apply to every agent. If this project could succeed, then moral objectivity would be guaranteed. Clearly, Kantians could compare moral positions on principled grounds and evaluate them irrespective of their personal perspective. However, this depends on the conceivability of logically deriving universal and categorical principles that entail reasons for a certain inter-personal behaviour from uncontroversial, necessary principles (Korsgaard 1998). Following Gibbard (1999: 145), I doubt that the most basic human value judgements are logically necessary though I cannot argue for it here. But if Kantian constructivism is understood as a theory about coherence, then it becomes a Victim Theory, too.

25Imagine emperor Caligula, who “aims solely to maximise the suffering of others. That is a horrendous life policy, but it needn’t be formally inconsistent” (ibid.). This is problematic for Kantians, but not because it is an issue about disagreement. Kantians might recognise Caligula’s policy as consistent, but still disagree with it; “we are coherent to do so” (ibid.). The disagreement is understood as follows: for us, minimising the suffering of others is consistent with our values; if someone of us utters “torture is wrong,” then this is true because it is consistent with our remaining normative beliefs. We disagree with someone like Caligula because we have very different basic values from which our value system is built. For Kantians, Caligula’s position on torture cannot be criticised any more than they can criticise his taste of ice cream. They might certainly have a different outlook than Caligula, but they cannot claim that he is fundamentally mistaken, independently of his or their personal outlook. Similar to sophisticated expressivism, Kantian constructivists have to jettison their objectivist aspirations. What is right or wrong depends ultimately on one’s personal outlook, at least if the logical deduction of normative values does not succeed.

3.3. Non-Robust Realism

  • 17 Which distinguishes the view from robust moral realism, which relates the truth and falsity of mora (...)

26Expressivism and Kantian constructivism (at least on some readings, cf. Gibbard 1999) are non-cognitivist positions. They entail that there are no moral facts and properties. However, the anything goes challenge seems to beset cognitivist metaethical positions, too. Reductionist forms of moral realism analyse the truth and falsity of moral sentences in virtue of causally efficacious nonmoral facts.17 The reliance on non-moral facts gives reductionists naturalistic credibility and promises that epistemic worries about moral knowledge do not arise. They also suggest that there will be a fact of the matter that decides whether the likes of Caligula are mistaken. Reductionists have clear objectivist aspirations. Popular varieties of naturalist realism, such as Brink (1989), Boyd (1988), as well as functionalists like Jackson (1998), Copp (2007), and Foot (2001) are representatives of this view. Reductionist forms of realism, however, are Victim Theories, too.

  • 18 Because if the epistemic worry goes through, then reductionism fails at the level of robust moral r (...)
  • 19 Which could have been, as Barkhausen argues in his paper.
  • 20 For reductionists, there are additional (and well known) worries related to the naturalistic fallac (...)

27The victimhood of reductionism has to do with nearby possible worlds. In a recent article, for instance, Barkhausen raises a dilemma for reductionists that arises out of their stance on relation of truth-conditions of moral sentences and evolution (Barkhausen 2016). The first horn of Barkhausen’s argument raises a worry about epistemic reliability similar to the one that confronts robust realism; it need not concern us here.18 The problem with objectivity arises on the second horn. The reductionist “allows that what is right or wrong […] covaries with our contingent evolutionary history” (ibid.: 22). The consequence is that, had evolution been different,19 the truth conditions of moral sentences would be different, too. Reductionism of this form “entails a morally repugnant relativism that is at odds with the verbal behaviour of ordinary moral reasoners. Most of us are deeply opposed to the idea that any way of coordinating on mutually beneficial behaviour that our moral evolution might have led us to endorse is as good as any other” (ibid.). In Barkhausen’s world, we meet the AGC’s repugnant characters of previous sections again. For example, slave-owners that cherish slavery, and slaves that do so, too. The reductionist allows that what they do is morally right. Again, disagreement can be accounted for by explaining why we believe slavery to be wrong while others deem it right. Yet, anything goes as far as it is conceivable to be a product of evolutionary processes.20

3.4. Conceptual Pluralism

28Finally, conceptual pluralists, another cognitivist account, is an “in-between notion of objectivity” (Timmons 1991: 379) that goes beyond culturally accepted standards of communities, but does not go all the way to assert fixed standards that have metaphysical backing. Putnam provides an example of conceptual pluralism (Putnam 1981). Conceptual pluralism avoids the epistemic troubles of robust realism and it is designed to “somehow carve out a non-absolutist conception of objectivity that avoids relativism” (Timmons 1991: 379). Putnam argues that moral claims have their truth-values relative to the speaker’s conceptual scheme. Additionally, he advocates a criterion of coherence of moral beliefs (Putnam 1981: 54-5). Conceptual pluralism avoids worries about the naturalistic fallacy, which might be relevant for reductionists, while it promises to show how moral claims can be objectively evaluated and compared. However, conceptual pluralism is a Victim Theory, too.

  • 21 Though see Horgan & Timmons 1991.

29In the case of Putnam’s realism, we do not even need to look to nearby possible worlds to see the challenge arise.21 Rational Nazis (if there were such a thing) in our midst are enough to highlight its problems. Putnam’s account essentially demands that normative claims are in coherence with one’s conceptual scheme. The problem arises in similar ways as in the cases of expressivism and constructivism. It is conceivable that one believes that torture is wrong, or that hurting animals is permissible, or that racism is good, while maintaining without contradiction that this belief is in coherence with all other beliefs of some suitable conceptual scheme. That is, given odd desires, conceptions, and beliefs, one might think that being a Neo-Nazi is a rational course of action. Depending on our conceptual schemes, Putnam can make good sense of why we disagree with such a person. But this is not enough, it seems. Even though Putnam allows for a plurality of moral codes, it seems that he must also show why “not just any conception of human flourishing is rational” for otherwise “we lose the idea that there is any substantive moral truth, and we lose our grip on how a morality or set of moral principles can be objective” (Timmons 1991: 386). Lacking an objectivist grip on moral truth, Victim Theorists can at best point to their principles of arriving at normative positions and hope that what others come up with jibes with their personal outlook. If not, then the Victim Theorists can surely explain why they disagree. But they are barred from comparing their normative outlooks in the relevant sense. As before, the VT has to allow that, ultimately, anything goes.

4. … Or Their Criterion for Moral Truth is Smug

30The previous section showed that all examined VTs face a variant of the following schematic problem:

i. According to the VT, agent A under circumstances C has most reasons to do Z.

ii. Z is appalling according to our (your and my) intuition.

  • 22 Though I am sceptical about the actual extend of moral disagreement, cf. Klenk forthcoming.
  • 23 I thank an anonymous reviewer for pressing me to address this point.
  • 24 That is, in accordance with the procedural or definitional criteria for determining normative posit (...)

31All VTs are committed to i and could explain ii; that is, they could explain just why we happen to think otherwise and thus give an account of what our moral disagreement consists in. For example, VTs may offer a moral psychology that explains why we form the beliefs and moral judgements that we do and thus explain why moral disagreement may appear like a foundational fact of our moral lives.22 They may furthermore suggest that any disagreement about i, or about whether or not to do Z, is resolved on first-order grounds. After all, they can explain ii and also why there may be good reasons for them to take ii seriously. So, as a matter of fact, moral disagreements appear settled, for the participants to the dispute. Obviously, however, this is a psychological or phenomenological point, and not a normative one.23 VTs cannot account for moral disagreement in the relevant sense. They cannot, reject i on grounds shared by their interlocutors. Specifically, they are unable to acknowledge and account for overarching reasons for or against Z that apply to all parties of the dispute. This is precisely the point made above: There is a sense in which you and I may find that Caligula lacks reasons to torture, but there is no sense in which these reasons apply to Caligula. Thus, VTs invite the view that their views legislate normative positions as well-formed24 that we find appalling. The baseness of the characters employed by the AGC is not the fundamental problem, however. It just indicates with a great vividness that the ultimate criterion for evaluating or comparing normative claims is an evaluator’s personal outlook. In that sense, all VTs turn out subjectivist theories that allow that, ultimately, anything goes, despite their objectivist aspirations.

32At this point, in respect to ii in particular, we face a decision. We either accept that i is the best that we can do (this seems to be Street’s position, for instance, Street 2008) and that ii is just a bullet that we have to bite. Apart from explaining where the disagreements between our normative position and that of some odd character lie, there is nothing further that the metaethical theory can do to retain a sense of objectivity. Putnam, for instance, considers this option. Citing Wittgenstein, he concedes that when confronted with certain odd characters, his justification has reached “bedrock, and this is where my spade is turned” (Putnam 1991: 85). Taking this route means to give up on even the minimal requirement of objectivity described earlier in this paper.

33Most VTs want to achieve more, however. Apart from giving an explanatory account of disagreement (as suggested above, VT may be well placed to explain what the disagreement consists in and why it arises), these theories want to license evaluation and comparison of radically conflicting moral claims and thereby establish that not anything goes. To do so, VTs have to be smug. In one sense, Victim Theorists can claim that Caligula ought to do things differently, or that he has reasons to change his ways. But ought and reasons in this sense are relative to the thinker of the thought and not to be interpreted from the perspective of Caligula. Consider Harman’s distinction between 1) the idea that it ought to be that an agent does X and 2) the idea that an agent ought to do X. Harman writes:

The first thought is the thought that there are reasons for wanting the agent to do it, which for us, the thinkers of the thought, the critics, are reasons that we have to want this to happen. The second thought is the thought that there are reasons the agent in question has to do that thing. These are quite independent thoughts. (Harman 1983: 317)

34Adherents of the Victim Theory may demand that it ought to be that all those troublesome characters like Caligula do differently, but their “ought” will be legitimate only in the first sense of Harman’s distinction. For example, that Caligula ought not to torture people for fun, that there are reasons to stop doing it, amounts to the claim that I have reasons to condemn Caligula’s behaviour. Issuing an “ought” in the second sense seems to require an agent-independent standard or an account of moral error that is applicable in cases of the AGC. Note that the way in which Caligula came to believe that torture is fun is not at issue. What we want in cases that illustrate the AGC is to be able to say that Caligula himself has reasons to do otherwise, not because of some other commitment that he accepts (which would point to the fact that Caligula’s set of beliefs is not coherent) but rather because of a consideration that everyone has decisive reasons to accept.

35It will be worthwhile to point out that practical consequences are not at issue here. That is, the desiderata for a solution to the AGC cannot include the practical consequence on the motivational reasons of agents. The urgency with which the AGC is sometimes directed at non-realists may leave the impression that practical consequences are indeed what is at issue. The thought seems to be that if a VT cannot explain why Caligula himself has reasons not to torture people, then we should soon expect Caligula-types are swarming the streets. This is not something that we should demand of a solution to the AGC. Again, the salience of ii) is that it implies a stark contrast with values that we presently hold, but we can abstract away from this normative nuisances to see that the ultimate agent-dependence of the Victim Theory is the cause of the AGC.

36Back to the objectivist aspirations of VTs. It seems that some VTs at least try to bluff their way to an account of moral error that provides the basis for calling Caligula out and claiming that he ought to do otherwise. Consider Gibbard’s expressivism as an example (Gibbard 2011). Gibbard repeatedly states that the wrongness of hurting a chicken is agent-independently wrong (ibid.: 40, 46). Call the fact that chicken and other animals feel pain CX. However, since the bare descriptive fact that animals feel pain does nothing on Gibbard’s theory, it must be that the wrong-making features of hurting an animal depend on someone taking CX to be a reason that counts against hurting animals. In the confrontation with animal tormentors, Gibbard realises the following situation:

Them: animal tormentors do not take CX to be a reason. So, we get N2, namely that CX is not a reason to stop activities that hurt the chicken.

Me: I (Gibbard) take CX to be a reason. So, we get N1, namely that CX is a reason to stop activities that hurt the chicken.

37I assume that we are with Gibbard on the issue of taking CX to be a reason against hurting animals. To avoid the first horn of the AGC, we must assume that there is something wrong with N2. Indeed, this is what Gibbard asserts (2011: 46). There is no fault in their reasoning, and no fault in our reasoning, yet we end up having different normative outlooks. The only reason to prefer our outlook, and declare it right as opposed to in error, is that it is our outlook. Victim Theorists that want to live up to their objectivist aspirations thus face a dilemma. They frustrate these aspirations by allowing that anything goes, or they are smug by making the criterion for moral truth and falsity depend on their personal outlook.

38It might turn out that this dilemma surrounding moral objectivity is a result of rejecting robust moral realism. For other reasons, error theorists insist that although morality appears to us in a certain way, we may just have to let go of certain desiderata, such an objective criterion (Joyce 2001, Mackie 1977). We need not give up, however, if an objective criterion for moral truth and falsity is implied by pragmatism.

5. Pragmatism is Neither Subjectivist Nor Smug

39Pragmatists have a response to the AGC that is neither subjectivist nor smug. Pragmatism is not a Victim Theory because its account of moral truth entails that a person-independent fact determines the truth-value of a moral claim.

40Pragmatism is committed to a strong form of fallibilism and consequently rejects certainty. Whether the nature of an animal’s pain counts as a reason against hurting an animal can thus never be asserted with certainty. No matter how good a belief might appear to us, no matter how good it would fit with our other beliefs, or how well it would fit the available evidence, the pragmatist will acknowledge that the belief could still be false (Misak 2002: 55). Importantly, however, pragmatists do not doubt that there is an independent reality – they only doubt that we could have reliable knowledge of it. This latter commitment plays and essential role in the pragmatist’s reply to the AGC.

41Let me explain what truly marks a difference between non-realism and pragmatism when it comes to the anything goes challenge. The fundamental difference is that the pragmatist is committed to the existence of facts that hold independently of a particular purview to serve as the arbitrator in an effort to evaluate and compare our reasons for action. Unlike non-cognitivist Victim Theories, who deny that such truths exist, and realist Victim Theories, who hold that a multitude of such truths may be on par, the pragmatist can maintain that truth is invariant in the relevant sense.

42Often, of course, pragmatism seems most plausibly connected to a naturalist position that would consider the truths to be conditional. For example, conditional on some evolutionary trajectory. But that is not necessarily the case, as illustrated by the variant of Peircean pragmatism discussed in this paper. Whatever will be favoured by our experiences in the limit may be unique and invariant, and commitment to that claim is what opens up a pragmatist rebuttal of the anything goes challenge. For only then can we maintain that Caligula, or the animal tormentors, or an internally coherent racist must be wrong on the grounds that determine moral truth for all.

  • 25 Naturally, it is a challenge for pragmatists to give concrete substance to the idea of a test again (...)

43Pragmatists can then analyse any given moral disagreement as a clash of experimental or indeed experiential hypotheses as to what would stand the test of experience. Moral disagreements are genuine disagreements about facts, on the pragmatist picture, and they allow criticism of one’s interlocutors based on shared, non-smug grounds. A robust realist position would prompt us to ask how we can ascertain the moral truth, and why any of the interlocutors can lay claim to moral insight in light of the disagreement, raising well-known challenges about moral disagreement. Pragmatism advances the debate because it suggests that our moral judgements can be understood as formulating hypotheses for experiments in living, or moral experiments (Mill 1869; cf. Klenk & Van de Poel 2021). The pragmatist can thus hold forth that the independent truth exists, that it is constituted by experience in the limit and, therefore, that it is an open question whether we or they are right. The claim that we are right is thus to be interpreted as a hypothesis that doing it our way, not tormenting the animals, will better accord with experience in the long run. That is ultimately an empirical claim to be tested against experience.25 In the confrontation with animal tormentors, Caligula, or whomever, who hold that N2 (to stand for some abject normative belief), the pragmatist can assume that there is a fact of the matter whether N2 or rather N1 would best fit our experience, were we to inquire as far as we could on the matter. In other words, for the pragmatist either N2 or N1 will be correct and the truth of this claim does not depend on the pragmatist’s personal outlook, but on contrasting both beliefs with our experience in the limit. Pragmatism also yields a procedural demand for pragmatists to follow. N2 would be false if we would find that we would still be in doubt about N2 were we to inquiry as far as we could on the matter.

44Consider Putnam’s rational racist. The racist holds a coherent belief set and believes N2. You and I believe N1. Pragmatism can pronounce one position to be mistaken, the other to be correct. The Nazi’s position would not stand up to experience would we inquire as far as we could into the matter. Why am I, the pragmatist, licenced to say this? Well, my experience matters on this issue. Moreover, it seems to me that somebody going after my life is an unpleasant experience. I imagine, then, that a belief with racist content would not “pass the test of experience” were we to inquire as far as we fruitfully could on the matter.

  • 26 It might be doubted whether the pragmatists, with their practical aspirations, should be concerned (...)

45Therefore, the pragmatist criterion for truth or falsity is objective. One test for objectivity is the independence from subjective preferences. This requirement is fulfilled, for my subjective experiences contribute to the test of experience, but they do not determine it. Furthermore, the pragmatist notion extends beyond inter-subjectivity because no agreement amongst agents is required. Rather, pragmatism makes room for moral truth and falsity in all circumstances because there is a fact of the matter whether N1 or N2 would pass the test of experience. All the pragmatist needs to presuppose is that beliefs are indeed susceptible to experience and that experience is somehow linked to how the world is. The pragmatist need never claim that our beliefs reflect how the world actually is, but only that our experiences of the world as we experience it is our best guide in our inquiry. Then the claim is that the way the world is determines the experiences that we have of it and that there is an aggregate of all experiences. Those beliefs that would best fit this hypothetical aggregate of experiences are assumed to be true. The troublesome characters introduced by the AGC can be handled in similar fashion. Were we to inquire as far as we fruitfully could on the matter, it would eventually turn out that either Gibbard is right or the animal tormentors. We would find that Caligula’s belief about the rightness of torture would succumb to the test of experience and that valuing each other’s dignity would win out. The pragmatist has one crucial advantage over the Victim Theories of his rivals: the pragmatist can legitimately assert that Caligula ought not to torture people, that the Nazi’s should not be racist, and so on,26 because the pragmatist’s criterion for assessing the truth and falsity of moral claims is ultimately independent of one’s own outlook.

46Some pragmatists might not be content with an account that relies on a hypothetical aggregate of experiences as a test for truth. One obvious problem is epistemological. If both the animal tormentors and I imagine whether N1 or N2 would pass the test of experience, we might plausibly arrive at different results. The animal tormentors will think of eons of generations of additional animal tormentors that have a great time while hurting animals, while Gibbard, you, and I will think of many more generations of cat-loving philosophers. Neither of both camps can claim special epistemic access to whatever would best fit our holistic experiences at the end of inquiry. This is true. But it may not be a problem in the context of answering the AGC, which was to supply an objective criterion to determine the truth and falsity of moral claims. All VTs that I discussed lack such a criterion, and some make good for it by being smug. Pragmatism, for that matter, need not rely on such means.

47In lieu of knowing what belief would best fit our experience, pragmatism is committed to a methodological principle and what might be called a restriction of their philosophical aspirations. The methodological principle is that genuine moral inquiry must proceed by continuously holding our beliefs to be responsive to new arguments (Misak 2002: 102). Pragmatists have to restrict their philosophical aspirations insofar as they must reject, at any given moment, the possibility of certainty. The implication is that, faced with troublesome characters like Caligula or rational racists, pragmatists cannot assert with certainty that the troublesome character is mistaken. Practically speaking, moral truth and falsity is assumed to be out of reach altogether, hence the subjunctive mood in the definition of the pragmatist account of truth. The best that pragmatism can do, then, is to assume that a belief is true in the pragmatist sense, as long as it fits with our experiences. Confrontations with rational racists or Caligula in philosophical thought experiments will or will not cast doubt on our normative beliefs, but this is something that the pragmatist asks you to experience rather than to rule out in principle.

48At this point, it is worth to return briefly to the therapeutic aspiration of pragmatism mentioned in the introduction. We are in the market for moral objectivity for good reason. It allows us to evaluate moral claims as true or false and compare them as better, worse, or on par. In that vein, pragmatists like Misak and Heney suggest that our practices are geared toward truth. In light of the discussion of the anything goes challenge, the crave for moral objectivity also appears as a desire to maintain disagreements as a shared endeavour with one’s interlocutors. If we can maintain objectivity, then interlocutors with conflicting outlooks are not discredited as falling outside the conversation, nor is one’s own outlook smugly taken as the ground of normative judgement. But perhaps there is a therapeutic aspect to pragmatism nonetheless. For it shows how we need not provide an account of the moral truths in a substantial sense to maintain moral objectivity. Instead, we may understand our moral judgements as experimental hypotheses about a shared world that guide our decision making, which does not presuppose access to the moral truth right now.

6. Objections

49I point out three possible objections against pragmatism. All three relate to the notion of a “test of experience” that is central to the pragmatist account of truth. The first objection questions the pragmatist reason for including all experience as relevant for inquiry. Why, Caligula might think, does it matter what his victims think about torture? Caligula might be convinced that his experience is all that matters and if a belief does not cause any doubt for Caligula, then it can be assumed to pass the test of experience. Fixing the range of relevant experience is to determine who and what counts as part of “our” experience that is relevant to determine the truth of normative beliefs.

  • 27 The matter ultimately depends on deeper questions about the division and interrelation of first-ord (...)

50Pragmatists can reply that their notion of truth is a bona fide metaethical claim that is not up for first-order moral discussion. Caligula may well think otherwise, as a matter of psychological fact, but he would simply be mistaken.27 That is, restricting the range of experience in this manner does not cause an issue for pragmatism. The validity of the criterion for objectivity is left intact, since it depends on analysis of the nature of assertion. Insofar as certain agents are interested in making genuine assertions, they should be prepared to expose their assertions to the greatest volume of experiences. What if a bunch of racist pragmatists adopts the pragmatist notion of truth and declares that truth is whatever would pass the test of experience of all that fit within their racist criteria? Such thought experiments do not threaten the pragmatist criterion of moral truth. If one accepts that a true belief is one that would not be doubted, then it will be relevant, even for the racist pragmatists, to have beliefs and make assertions that would not cause themselves to doubt them. For example, if they were to assert things that most of us find morally base, then psychological, sociological, and eventually biological processes will frustrate any attempt to consider only their own experiences as relevant for the truth.

51A more pressing issue concerns the epistemic access to a hypothetical aggregate of all experiences. If pragmatism indeed links its account of truth to whatever the aggregate of experiences and arguments, then an epistemological worry similar to the reliability challenge against robust moral realism (cf. Klenk 2020) might arise. The notion of, roughly, “all experiences ever,” E, turns into a fact against which the pragmatist measures beliefs and assertions. Accordingly, N1 and N2 are true only if they match E. The claim that “N1 matches E” is reminiscent of robust moral realism and the conditions for knowing its truth-conditions are comparably vague. This is a serious worry for pragmatism because of a well-known problem for stipulating facts that are epistemically inaccessible, which goes back to Quine (see Hopster & Klenk 2020). That challenge will be decided on general, epistemological grounds. For present purposes, it is sufficient to note that the pragmatic interpretation of our moral judgements as conveying experiential hypotheses may supply one of the reasons that legitimises the stipulation of mind-independent and epistemically inaccessible truth.

52Third, a worry might be that pragmatism is, indeed, a form of constructivism and thus relies on a notion of agreement in determining the truth-value of normative claims. How else, critics might ask, could pragmatists settle whether N1 or N2 does, at this moment, best fit our experiences other than in some (real or hypothetical) discourse procedure? If actual agreement would be what counts, rather than a match with the hypothetical notion of aggregate experiences E, then the worry about the range of experience would also return, for it would not be clear what the pragmatist can say if faced with a whole community of rational racists, frighteningly unison in their relishing of racist acts.

53But the pragmatist might have a pragmatist answer to this, roughly along the following lines. First, there are good arguments to the effect that we cannot have certainty (i.e. know about the mind-independent world). Second, given this restriction, we have to accept that there are no immutable standards that we can know of. Third, given this insight, the pragmatist methodology is the best that we can follow (which is also based on the hope that beliefs that are not challenged by doubt is our best guiding light). Fourth, if following this methodology will eventually lead to us believing that animal pain does not matter, then this is the best that we could do.

54In light of these worries, it should be clear that difficult questions remain before we can confidently proclaim pragmatism as a solution to long-standing metaethical issues to do with moral objectivity. For example, about how to aggregate experiences, and which or whose experiences to take into account in determining truth. Consequently, this article does not vindicate pragmatism as the best account of a type of moral objectivity worth having. What I do hope to have shown, however, is a novel and interesting sense in which pragmatism goes beyond rival Victim Theories as well as robust realism, carving out an interesting position that may offer all the moral objectivity worth wanting. Future work should be directed both at fleshing out the contours of such a pragmatist account, and to further the comparison to a robust realist alternative.

7. Conclusion

55Pragmatism is rarely discussed in recent metaethical debates. In particular, it is not recognised as a probable solution to the pervasive anything-goes challenge to non-realist metaethical theories. Non-realist theories are unified in their rejection of robust moral realism (for various reasons) and their aspiration to preserve a sense of objectivity in ethics. However, I argued that non-realist theories turn out to be Victim Theories: they fall prey to the anything goes challenge. They either give up on their objectivist aspirations or retain a minimal sense of objectivity by being undesirably smug. I have argued that Peircean pragmatism solves this aspect of the anything goes challenge by providing an objective criterion for moral truth and falsity. I have also argued that pragmatism faces problems that relate to the details of its account of moral truth. Apart from these worries, pragmatism’s structural advantage over Victim Theories gives us a reason to consider it as genuine metaethical answer to the arguable shortcomings of robust realism.

Top of page


Ayer Alfred Jules, (1971 [1936]), Language, Truth, and Logic, Harmondsworth, Middlesex, Penguin Books.

Barkhausen Max, (2016), “Reductionist Moral Realism and the Contingency of Moral Evolution, ”Ethics, 126 (3), 662-89.

Benacerraf Paul, (1973), “Mathematical Truth,” The Journal of Philosophy, 70 (19), 661-79.

Blackburn Simon, (1998), Ruling Passions: A Theory of Practical Reasoning, Oxford, Clarendon Press.

Boyd Richard, (1988 [1995]), “How to Be a Moral Realist,” in Geoffrey Sayre-McCord (ed.), Essays on Moral Realism, Ithaca, Cornell University Press, 181-228.

Brink David O., (1989), Moral Realism and the Foundations of Ethics, Cambridge, Cambridge University Press (“Cambridge Studies in Philosophy”).

Chang Ruth, (2002), “The Possibility of Parity,” Ethics, 112 (4), 659-88.

Copp David, (2007), Morality in a Natural World: Selected Essays in Metaethics, Cambridge, Cambridge University Press.

Darwall Stephen L., Gibbard Allan & Peter Railton, (1992), “Toward Fin de siecle Ethics: Some Trends,” The Philosophical Review, 101 (1), 115-89.

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

Field Hartry, (1980), Science without Numbers: A Defence of Nominalism, Princeton, Princeton University Press.

Foot Philippa, (2001), Natural Goodness, Oxford, New York, Oxford University Press.

Gibbard Allan, (1990), Wise Choices, Apt Feelings: A Theory of Normative Judgment, Cambridge, Mass., Harvard University Press.

Gibbard Allan, (1999), “Morality as Consistency in Living: Korsgaard’s Kantian Lectures,” Ethics, 110 (1), 140-64.

Gibbard Allan, (2003), Thinking How to Live, Cambridge, Mass., Harvard University Press.

Gibbard Allan, (2011), “How Much Realism? Evolved Thinkers and Normative Concepts,” in Russ Shafer-Landau (ed.), Oxford Studies in Metaethics: Volume 6, Oxford, New York, Oxford University Press.

Habermas Jürgen, (1991), Erläuterungen zur Diskursethik, Frankfurt am Main, Suhrkamp, (“Suhrkamp Taschenbuch Wissenschaft”).

Harman Gilbert, (1983), “Human Flourishing, Ethics, and Liberty, ”Philosophy & Public Affairs, 12 (4), 307-22.

Heney Diana, (2016), Toward a Pragmatist Metaethics, London, Routledge.

Hills Alison, (2010), The Beloved Self: Morality and the Challenge from Egoism, Oxford, Oxford University Press.

Hopster Jeroen & Michael Klenk, (2020), “Why Metaethics Needs Empirical Moral Psychology,” Critica, 52 (155), 27-54.

Horgan Terence & Mark Timmons, (1991), “New Wave Moral Realism Meets Moral Twin Earth,” Journal of Philosophical Research, 16, 447-65.

Jackson Frank, (1998), From Metaphysics to Ethics: A Defence of Conceptual Analysis, Oxford, Clarendon Press.

Joyce Richard, (2001), The Myth of Morality, Cambridge, New York, Cambridge University Press (“Cambridge Studies in Philosophy”).

Klenk Michael, (2020), “Third Factor Explanations and Disagreement in Metaethics,” Synthese 197, 427-46. Online:

Klenk Michael, (2022), “Debunking, Epistemic Achievement, and Undermining Defeat,” American Philosophical Quarterly, 59 (1).

Klenk Michael, (forthcoming), “Moral Realism, Disagreement, and Conceptual Ethics,” Inquiry.

Klenk Michael & Ibo van de Poel, (2021), “COVID-19, Uncertainty, and Moral Experiments,” History and Philosophy of Life Sciences, 43 (1), 8.

Korsgaard Christine M. (ed.), (1998), The Sources of Normativity, Cambridge, Cambridge University Press.

Lekan Tood, (2003), Making Morality. Pragmatist Reconstruction in Ethical Theory, Nashville, Vanderbilt University Press.

Mackie John L., (1977), Ethics: Inventing Right and Wrong, London, Penguin Books.

Mill John Stuart, (1869), On Liberty, London, Longman, Roberts and Green.

Miller Alexander, (2007), An Introduction to Contemporary Metaethics, Cambridge, Polity Press.

Misak Cheryl J., (2002), Truth, Politics, Morality: Pragmatism and Deliberation, London, New York, Routledge.

Misak Cheryl J., (2013), The American Pragmatists, Oxford, Oxford University Press (“The Oxford History of Philosophy”).

Peirce Charles Sanders, (1935-1958), The Collected Papers of Charles Sanders Peirce, vol. I-VI, ed. P. Weiss, C. Hartshorne, vol. VII-VIII, ed. A.W. Burks, Cambridge, Mass., Harvard University Press.

Putnam Hilary, (1981), Reason, Truth, and History, Cambridge, New York, Cambridge University Press.

Putnam Hilary, (1991), Representation and Reality, Cambridge, Mass., The MIT Press.

Railton Peter, (1986), “Moral Realism,” The Philosophical Review, 95 (2), 163-207.

Rawls John, (1980), “Kantian Constructivism in Moral Theory”, The Journal of Philosophy, 77 (9), 515-72.

Shafer-Landau Russ, (2003), Moral Realism: A Defence, Oxford, Clarendon Press.

Smith Michael A., Lewis David & Mark Johnston, (1989), “Dispositional Theories of Value,” Proceedings of the Aristotelian Society, Supplementary Volumes 63, 89-174.

Street Sharon, (2006), “A Darwinian Dilemma for Realist Theories of Value,” Philos Stud, 127, 109-66.

Street Sharon, (2008), “Constructivism about Reasons,” in Russ Shafer-Landau (ed.), Oxford Studies in Metaethics: Volume 3, Oxford, New York, Oxford University Press, 207-45.

Timmons Mark, (1991), “Putnam’s Moral Objectivism,” Erkenntnis, 34 (3), 371-99.

Wielenberg Erik J., (2014), Robust Ethics: The Metaphysics and Epistemology of Godless Normative Realism, Oxford, Oxford University Press.

Wiggins David, (2002), “Truth, and Truth as Predicated of Moral Judgements,” in David Wiggins (ed.), Needs, Values, Truth: Essays in the Philosophy of Value, Oxford, Clarendon Press, 139-84.

Wilson Catherine, (2010), “Moral Progress Without Moral Realism,” Philosophical Papers, 39 (1), 97-116.

Top of page


1 The pragmatism that I’ll discuss here is thus in stark contrast to Rorty’s nihilist view, which is often associated with the pragmatist label. Rorty 1998 abandons the notion of truth and instead focuses on justifiability amongst a group of inquirers. Rorty’s view thus waves goodbye to objectivity and, consequently, falls out of the scope of this paper.

2 However, see Wilson 2010 for a recent attempt.

3 The challenge relates to the challenge of rational egoism in first-order normative ethics: Thrasymachus, Hume’s Knave, etc. See Hills 2010.

4 By “ultimately” I mean that the final answer to repeated iterations of “why” questions about moral claims is not determined by one’s personal outlook.

5 They might also be “on par,” see Chang 2002. In any case, the relevant criteria (metaphysical, conceptual, etc.) for assessing the trichotomy is another matter. My criterion is just the plausible one that a metaethical theory should be able to theoretically underwrite this way of talking about morality.

6 A third option is that VTs abandon the project of accounting for moral truth in some sense. The resulting nihilist views, however, are not within the scope of this paper.

7 In between realist objectivity and subjectivism.

8 Henceforth I use pragmatism to denote the pragmatism championed by Misak in the spirit of Peirce. The accounts of truth and moral objectivity by the other classical pragmatists Dewey and James are markedly different. Neo-pragmatists such as Quine, Rorty, and Sellars also deserve a nuanced treatment that would get me too far into interpretative and comparative terrain than necessary.

9 Thus, this also applies if one is trying to mislead with a false assertion. Lying, for instance, depends on taking the other’s claim to be sincere and thus aim at truth (Misak 2002: 59-60).

10 A belief aims at truth iff I believe p, I believe p to be true (Wiggins 2002). If this is right then p must be sensitive to something and the pragmatist stresses its sensitivity to human experience.

11 There might be worries about lying and bullshitting in Frankfurt’s sense and Misak does not address these. However, the use of the qualifier “proper” in the argument might help to avert such issues.

12 “Logic inexorably requires that our interests [and experiences] shall not be limited. They must not stop at our own fate, but must embrace the whole community” (Peirce CP 5.354, cited in Misak 2013: 37).

13 Recall that non-sophisticated expressivism such as that of Ayer (1971 [1936]) has no ambition to retain objectivist seeming so it falls out of the scope of this paper.

14 It might be questioned whether the anything goes challenge is worth answering. There are two replies. First, it is relevant based on the extend one desires to preserve objectivist pretentions. This ties in with the second answer: many scholars seem concerned about the AGC, in response to quite different non-realist positions.

15 Except, perhaps, one frees oneself from the deterministic course of events and thus achieves Kant’s freedom – but I doubt that this is what Gibbard wants to maintain.This might also raise a problem about “ought implies can” – Gibbard will most certainly not stick to N1 were he in the shoes of the animal tormentors; he would adhere to N2 by definition!

16 Humean constructivism, as defended by Street 2008, is not a Victim Theory because it does not aspire to retain objectivity in ethics.

17 Which distinguishes the view from robust moral realism, which relates the truth and falsity of moral sentences to causally-inert moral facts.

18 Because if the epistemic worry goes through, then reductionism fails at the level of robust moral realism - even before it reaches the potential of becoming a Victim Theory.

19 Which could have been, as Barkhausen argues in his paper.

20 For reductionists, there are additional (and well known) worries related to the naturalistic fallacy.

21 Though see Horgan & Timmons 1991.

22 Though I am sceptical about the actual extend of moral disagreement, cf. Klenk forthcoming.

23 I thank an anonymous reviewer for pressing me to address this point.

24 That is, in accordance with the procedural or definitional criteria for determining normative positions within the respective VT.

25 Naturally, it is a challenge for pragmatists to give concrete substance to the idea of a test against experience. Contemporary pragmatists have proposed various ways to make the test of experience more concrete. For example, Heney’s 2016 recent account of a pragmatist metaethics explores how vindication that is rationally unimpeachable in practice may be a useful operationalisation of the experiential limit. Evaluating the details of this proposal, or others like it (e.g. Lekan 2003), is, however, beyond the scope of this essay.

26 It might be doubted whether the pragmatists, with their practical aspirations, should be concerned with the legitimacy of such theoretical notions in the face of racists and psychopaths, but that is another question and, in any case, applies to all other philosophical theories that I discussed, too.

27 The matter ultimately depends on deeper questions about the division and interrelation of first-order normative claims and metaethics, which is beyond what I can address here.

Top of page


Electronic reference

Michael Klenk, “Pragmatism and Moral Objectivity”European Journal of Pragmatism and American Philosophy [Online], XIII-2 | 2021, Online since 20 December 2021, connection on 15 June 2024. URL:; DOI:

Top of page

About the author

Michael Klenk

Delft University of Technology

Top of page



The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) are “All rights reserved”, unless otherwise stated.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search