1 Quasi-Realism and The Mind-Independence Claim

Expressivism is a metaethical position that explains the meaning of moral judgments indirectly by citing the mental states they are typically used to express, and what distinguishes it from metaethical cognitivism is that it denies that moral sentences express cognitive mental states. According to expressivism, we utter moral sentences to express non-cognitive mental states. For convenience, in this article I follow some other philosophers in calling them “conative attitudes”.

A major difficulty for expressivism is that the surface features of our moral discourse favor metaethical cognitivism and even realism. It is quite common for us to say that we believe something is morally right or wrong, our belief is true, and its being morally right or wrong is a moral fact that obtains independently of how we actually think and feel about it. Some expressivists attempt to explain the “realist-seeming” features of our moral discourse on an expressivist basis. This is why they name their metaethical position as “quasi-realism”. My major concern in this article is to assess whether quasi-realism can accommodate the notion of mind-independence embedded in our ordinary moral discourse.

Many objections to quasi-realism focus on whether the meaning of moral sentences can remain constant across asserted and unasserted contexts so that moral modus ponens can be valid, which is an important aspect of the Frege-Geach problem (Geach, 1965). Although quasi-realists have developed some strategies to handle this problem (Blackburn, 1984, pp. 189–196, 1993, pp. 182–197, 1998, pp. 70–73; Gibbard, 1990, pp. 83–102, 2003, pp. 41–87), there are still objections questioning whether we can ascribe the notion of logical validity to moral modus ponens within the framework of expressivism (Hale, 1986, 1993; Wright, 1988; Schueler, 1988). I will not attack quasi-realism from this direction. It is not clear whether we actually and must grasp moral modus ponens as logical inference, given that quasi-realists have already provided us with alternative stories about it and the intuition of ordinary moral speakers at least does not obviously count against those quasi-realist accounts. In this article, I specifically examine the quasi-realist interpretation of the mind-independence claim about morality. I believe the notion of mind-independence is embedded in our ordinary moral thinking in a distinctive way, the distinctiveness of which the quasi-realist account cannot explain. This is where expressivists cannot reconcile their metaethical commitments with our ordinary moral thinking.

The mind-independence claim about morality is not alien to ordinary moral speakers. Let us suppose that the claim to mind-independence embedded in our everyday moral discourse, in its general form, can be captured by the negation of the following counterfactuals (take x to be an object of moral evaluation and M to be a moral predicate):

(1) If I thought x is M, then x would be M.

(2) If I did not think x is M, then x would not be M. Footnote 1

If moral judgments express conative attitudes and therefore do not aspire to describe an independent moral reality, how should we make sense of the mind-independence claim that seems to have ontological implications?

Quasi-realists should answer this question in a way that can meet the challenge posed by the Frege-Geach problem in general. Schroeder (2008) has rightly pointed out that everything we can do syntactically with a non-moral descriptive term we can also do with a moral term, and moral terms have the same semantic properties as natural terms in complex sentences. The general worry beneath the Frege-Geach problem is that it is difficult to see how this could be the case if expressivism is true. Concerning specifically the mind-independence claim, since a counterfactual without moral terms entails a certain relation obtaining between its antecedent and consequent, a plausible expressivist account of (1) and (2) (and their negation) should also accommodate such a relation. Now we know that if we accept a counterfactual conditional, then we rationally ought to accept its consequent under counterfactual circumstances in which we accept its antecedent. Thus, it is not coincidental that expressivists generally give us an “internal” reading of the mind-independence claim, according to which the claim to moral mind-independence itself is a moral claim about how we ought to form moral commitments, and this seems like the only possible way to accommodate the relation obtaining between the antecedent and the consequent. To accept (1) and (2) is to accept that I should form a moral commitment under counterfactual circumstances in which I believe I have such a commitment, and reject a moral commitment under counterfactual circumstances in which I believe I do not have it. To deny (1) and (2) is to claim that I should not (Blackburn, 1981, p. 179, 1984, pp. 217–219, 1998, pp. 311–312; Gibbard, 1990, pp. 164–166, 2003, p. 183, p. 186; Sinclair, 2008, pp. 267–272).Footnote 2 The mind-independence claim construed by quasi-realists in this way is a moral claim regulating the way we form moral commitments. If quasi-realists are correct, then what the claim to moral mind-independence actually means is that we should not form moral commitments in response to factual beliefs about what moral commitments we have.

In this article, I argue against such a quasi-realist “internal” reading of the mind-independence claim. My general objection is that expressivists cannot explain why the majority of us strongly tend to interpret the mind-independence claim as a non-normative metaethical claim, even if the “internal” reading is correct and thus it is actually a moral claim. The first problem related to this objection is that if the internal reading is correct and ordinary moral speakers are at least implicitly aware of it, we can hardly explain why so many metaethical theorists, who are also competent moral speakers, read the mind-independence claim in a metaethical way. As a first-order moral claim, the internal reading of moral mind-independence is supposed to be neutral on all the metaethical semantic accounts and therefore has no substantive metaethical implications. The second problem related to it arises from the fact that ordinary moral speakers, at least in some non-philosophical contexts, also intuitively favor the metaethical interpretation of the mind-independence claim. However, as I point out, if expressivists would like to explain why the intuition of ordinary moral speakers sometimes also strongly counts against the quasi-realist internal reading, it is not clear whether they can avoid intolerably ad hoc postulations. I then consider a possible expressivist reply to this objection. Expressivists may argue that the mind-independence claim is a first-order moral claim that enjoys a conceptual status, and this is why it has some features that are typically possessed by metaethical claims. I argue that at least for expressivists, such an idea is of no help for them to meet the challenge I raise. Expressivists can at best claim that the acceptance of the mind-independence claim is constitutive of thinking and talking in moral terms, but our intuition against the quasi-realist internal reading does not seem to rely on whether competent moral speakers actually intend to engage in moral thinking or talking. This shows that in order to defend the quasi-realist account of moral mind-independence, there is still more work to be done.

2 The Metaethical Mistake

The first difficulty for the quasi-realist internal reading is to explain the metaethical mistake prevalent among other philosophers. The problem is that if the internal reading is correct, then we can hardly explain why the majority of metaethical theorists, who are also competent moral speakers, take the mind-independence claim as a non-moral metaethical claim.Footnote 3

To raise such a problem may sound question-begging. It is not clear whether people’s disagreeing with a metaethical claim is sufficient as a challenge to it. The meaning of moral sentences is never maximally transparent to competent moral speakers, which is why there are so many unsettled disputes in the domain of moral semantics. Then, why take other people’s objection to a semantic account as obvious evidence for its falsity? Moreover, if misinterpreting the mind-independence claim is a metaethical mistake that calls for an explanation, then both cognitivists and expressivists owe us a story about why the other side is mistaken, which means this is not a problem particular to expressivism.

I do not think this problem can be dismissed so easily. Admittedly, given the tight connection between moral judgments and motivation, it may not be transparent to us what mental states are exactly expressed by moral claims, which in turn makes expressivists and cognitivists divided on the meaning of moral sentences. However, it still sounds implausible to claim that we can have difficulty in distinguishing between a normative claim that is essentially action-guiding, and a non-normative descriptive claim that is not. To accuse people of failing to recognize such a distinction is to assume that competent moral speakers can sincerely make a moral claim about what practically matters to them, but somehow mistakenly believe that they are just making a non-normative descriptive claim that by itself has no practical significance, or vice versa. Given the intuitive absurdity of this accusation, such a mistake is supposed to be quite rare among us. This is why I believe cognitivists and expressivists both owe us an explanation about why their opponents are mistaken about moral mind-independence. Furthermore, I do not mean that this is exclusively a problem for expressivists. What I will demonstrate is that cognitivists are in a much better position than expressivists to give us a plausible explanation, which is a good reason for us to reject the quasi-realist account of moral mind-independence.

To remain charitable, in this section let us first assume that ordinary moral speakers are at least implicitly aware of whether moral mind-independence is a moral notion or a metaethical one in an everyday context. Presumably, the dispute over the mind-independence claim has its root in a philosopher’s mistake —— a mistake we are prone to when we distance ourselves from everyday moral discourse to critically reflect on it. Probably, it is the philosophical preconceptions of metaethical theorists that misguide them about the nature of the mind-independence thesis, especially when they are in an argumentative philosophical context. Of course, this is not to deny that even in a non-philosophical everyday context, a moral claim with a complex structure may still mislead us to interpret it as a metaethical one, or a complex metaethical claim part of which is a moral sentence may misguide us to read it as a moral one. In the next section, I consider such a possibility, and I argue that this possibility is even more challenging for expressivists.

Now we should assess which side can give us a better explanation of why the other side is mistaken about moral mind-independence. Given the assumption that people implicitly know the nature of the mind-independence claim in non-philosophical contexts, it is reasonable to expect that they would not misunderstand its status of being a moral or non-moral claim when there is no obvious evidence counting against their ordinary understanding of it. We need a plausible explanatory story about how a philosopher’s metaethical commitments could challenge their implicit belief about the nature of moral mind-independence.

Cognitivists typically take the mind-independence claim, when formulated as negated counterfactuals like (1) and (2), to be a metaethical claim denying moral facts counterfactually covary with facts of our mental states. Let us first assume that they are right. Surprisingly, quasi-realists instead tell us that talking about moral mind-independence is still moral talking, and this is the only intelligible way to talk about it (Blackburn, 1998, p. 311; Gibbard, 2003, p. 186). If quasi-realists are also competent moral speakers, why do they insist on such a misinterpretation? What might mislead them about what we are up to when talking about moral mind-independence?

The answer is obvious. The metaethical commitments of expressivists leave no space for moral facts (understood in a non-minimalist, metaphysical sense), let alone any dependency-relation between moral facts and facts about what moral commitments we have. If cognitivists are right, then the correct reading of the mind-independence claim simply cannot be incorporated into an expressivist framework. Philosophers who refuse to give up metaethical expressivism must find a way to reconcile moral mind-independence with their metaethical commitments. Then, how to interpret moral mind-independence in an expressivist-friendly way? They cannot construe it arbitrarily, because, as is already pointed out by Schroeder (2008), complex sentences with moral terms have the same sort of semantic properties as complex sentences using only non-moral descriptive terms. For conditionals like (1) and (2), a plausible account of their meaning must capture the relation a conditional sentence bears to its antecedent and consequent. We know that when accepting a conditional we also accept that we rationally ought to endorse its consequent in response to the endorsement of its antecedent. Now we can plausibly explain why quasi-realists interpret the mind-independence claim as a moral claim: Such an interpretation can ensure that the acceptance of a mind-dependence conditional and its antecedent rationally commits us to the acceptance of its consequent. Although this account seems to have blurred the boundary between epistemic normativity and practical normativity, at least it can allow the mind-independence claim to serve the function of regulating the way we form moral commitments in response to certain factual beliefs, just like the cognitivist account can do.

However, if the quasi-realist internal reading is correct, it does not seem like quasi-realists can give us an equally good explanation of why cognitivists are mistaken about it. Suppose that the mind-independence claim is actually a moral claim. Cognitivists certainly deny moral mind-independence should be understood in this way. They still insist that the mind-independence thesis is a metaethical thesis denying moral facts counterfactually covary with facts about the moral commitments we have. Given that cognitivists are also competent moral speakers, why are they so badly mistaken about such an ordinary notion embedded in our moral thinking?

Here we may find something puzzling. If the mind-independence claim is a first-order moral claim, then all sorts of metaethical semantic accounts are supposed to be neutral on it. Certainly, expressivism is perfectly compatible with it, but so is cognitivism. A second-order metaethical semantic theory neither confirms nor denies a first-order moral claim, but only tells us what we mean when we make this claim. If we assume competent moral speakers at least implicitly know that moral mind-independence is a first-order moral notion, then it is hard to explain why any metaethical semantic story would distort our understanding of it. A first-order moral claim would be compatible with any metaethical account of its meaning, thus no metaethical theorists will be pressured into denying its status of being a substantive moral claim (and endorsing a metaethical interpretation of this claim would make them even more vulnerable to the challenge of some versions of metaethical subjectivismFootnote 4). There is no need for cognitivists to misinterpret these counterfactuals as metaethical sentences to be consistent.

Furthermore, given that a first-order moral claim has no substantive implications on the metaethical level, it does not particularly favor those metaethical theories that entail the cognitivist metaethical reading of moral mind-independence. Suppose that the quasi-realist account of the meaning of this moral claim is correct, thus the problem of moral mind-independence is a moral problem about what features to look into in moral inquiry and what should be taken into account in a moral decision (Blackburn, 1993, p. 153; Gibbard, 2011). Anyone who endorses the mind-independence thesis will accept that their moral evaluation of an object should not be responsive to factual beliefs about their own moral commitments. However, this does not by itself imply whether there are moral facts (in a non-minimalist sense) that do not counterfactually covary with facts of our mental states. No matter what is the correct metaethical theory, it is always intelligible to make a first-order moral claim without implying we should accept any particular metaethical claim about its meaning.Footnote 5 It still looks puzzling why so many metaethical theorists mistake this moral claim as a metaethical one.

Therefore, if we assume competent moral speakers at least implicitly know that the mind-independence claim is a moral claim or a metaethical one in non-philosophical contexts, cognitivists are in a better position than expressivists to explain why their opponents can be mistaken about it. Quasi-realists still owe us a plausible explanation about why other metaethical theorists are susceptible to such an obvious mistake, and it sounds ironic to say only the quasi-realists are intelligent enough to avoid this mistake. Perhaps for expressivists, a more promising strategy is to claim that ordinary speakers sometimes tend to misunderstand the notion of moral mind-independence even in non-philosophical contexts. There are metaethical theorists grasping the mind-independence claim as a metaethical claim because this is a mistake embedded in our ordinary moral thinking and talking. This may explain why people are susceptible to this mistake even if all sorts of metaethical semantic accounts are neutral on substantive moral claims. In the next section, I assess whether such a strategy can make quasi-realists get off the hook.

3 The Ordinary Mistake

Now we should consider the possibility that the misinterpretation of moral mind-independence is not merely a philosopher’s mistake. Probably competent moral speakers, at least in some contexts, tend to misunderstand the mind-independence claim even if they are not engaged in philosophical reflection of it. Then, quasi-realists can argue that there are so many metaethical theorists falling prey to such a mistake because we as competent moral speakers sometimes simply do not know whether we are talking morally or metaethically in an everyday context.

There is good evidence that ordinary moral speakers at least sometimes take moral mind-independence to be a metaethical notion. Among people who have not been trained in moral philosophy, a common objection to metaethical expressivism is still that it makes morality objectionably mind-dependent, which implies a metaethical understanding of moral mind-independence (Olson, 2010). Therefore, it should not be surprising that sometimes our intuition does not favor an internal reading of mind-dependence counterfactuals like (1) and (2) (and their negation).

A thought experiment can make this intuition clear. Suppose that there is a group of people who accept the mind-independence claim as we do. One day a catastrophe radically reshaped their moral character, and they changed their moral views on everything. Now all of their moral commitments have been different from what they had before. However, as long as they still take their newly acquired moral views seriously, I think almost all of us (perhaps except quasi-realists) will intuitively believe that they still accept the general form of the mind-independence claim. Intuitively speaking, we believe they still agree that, for any object of moral evaluation x that is M, it is not the case that “if I did not think x is M then x would not be M”; for any x that is not M, it is not the case that “if I thought x is M then x would be M”.

It seems that many of us intuitively believe one can radically change one’s moral views on everything while retaining one’s commitment to moral mind-independence. This means sometimes we tend to intuitively deny that the mind-independence claim is a moral claim. Cognitivists would say we have this intuition because it really is a non-moral claim, and quasi-realists do not share this intuition because they are misguided by their metaethical commitments. However, quasi-realists must accuse us of failing to distinguish between a moral claim and a non-moral one in this case. The tricky point is that, as mentioned earlier, normally it is quite easy for us to recognize such a distinction, and the quasi-realists still owe us a story about why so many of us sometimes tend to intuitively deny that moral mind-independence gives us a moral requirement.

Now let us assume that moral mind-independence is a moral notion. Then, why do we have the misleading intuition that it is not? Quasi-realists should tell us what features of the mind-independence claim make us susceptible to misinterpreting it. I cannot predict what quasi-realists will say about it, but I will try to give some tentative answers on behalf of the quasi-realists with the hope of uncovering the most difficult part of meeting this challenge.

A possible explanation is that the mind-independence claim has what Cuneo & Shafer-Landau (2014) call the “framework status”, which fixes “the boundaries as to what counts as a type of subject matter” (p. 407).Footnote 6 If there are some people sincerely denying the mind-independence claim, we would suspect that their “moral” discourse is not about morality at all. We may think that their “moral” judgments do not express genuine moral views but personal tastes and preferences. This framework status may misguide us to believe that the mind-independence claim is a non-moral metaethical claim people always accept so long as they commit to moral thinking and talking, irrespective of what particular moral commitments they have.

However, this is insufficient to explain our supposedly misleading intuition. As noted by Cuneo & Shafer-Landau (2014), some other first-order moral claims, the truths of which are quite evident, also enjoy this framework status. If some people deny killing an innocent person or torturing a baby for fun is pro tanto morally wrong, it also looks dubious whether they are really talking about moral wrongness at all. Clearly, none of us will be disposed to treat these moral claims as non-moral ones.

Apart from having a framework status, there is one more thing that distinguishes the mind-independence claim from other ordinary moral claims. It comprises counterfactuals like (1) and (2) in which moral sentences are not assertoric. Can this explain our supposedly misleading intuition? Blackburn (1998) claims that conditionals with a non-moral antecedent and a moral consequent express our moral standards. Cognitivists can also agree with this. The difference is that quasi-realists identify these conditionals as moral sentences, while cognitivists would interpret them as metaethical ones. Nevertheless, no matter which side is correct, it is not counterintuitive that we would reject some highly-plausible moral standards if we significantly changed our moral views. Suppose that we reject the moral standard “If doing so killed an innocent person then it would be morally wrong”, which involves no assertoric use of moral sentence. Such a rejection may also make others suspect that we are not talking about moral wrongness at all, indicating the framework status of this moral standard. However, our rejection of such a moral standard would be perfectly intelligible if we endorsed a radically different moral outlook. This is quite different from our intuitive grasp of moral mind-independence: We intuitively believe that we can endorse completely different moral commitments without giving up our commitment to the mind-independence claim.

Admittedly, it is not clear whether we find rejecting such a moral standard intelligible because the moral standard itself is a moral claim, or because it is a metaethical claim the acceptance of which must rely on a particular moral view about “killing an innocent person”. What is clear is that we intuitively agree changes in moral outlook can co-occur with changes in the moral standards we accept, but deny changing moral outlook will result in changing our view on moral mind-independence. This means there is still an intuitive difference between a standard-expressing conditional and the mind-independence claim: Our intuition explicitly counts against the latter being a moral claim but not the former. The fact that the mind-independence claim enjoys a framework status and involves no assertoric moral sentences may make it look like a metaethical claim, but it is still not sufficient to explain why our intuition explicitly denies its status of being a moral claim.

I do not mean that quasi-realists cannot give us a satisfactory explanation, but I think it has been made salient how difficult such a task is. The problem underlying these difficulties is that the mind-dependence counterfactuals and their negation seem to enjoy a fundamentally different status from other counterfactuals with a similar form. Consider the schema “For any x that is not M, if x had the natural property N, then x would be M; for any x that is M, if x did not have the natural property N, then x would not be M”. How do we decide whether to accept or reject it given a particular N and M? When the natural property N is “being judged as M”, most of us would reject it. We deny that whether x is M is counterfactually dependent on whether we judge it to be M, and as indicated above, our intuition suggests that we reject it irrespective of what particular moral commitments we have. However, when the natural property N is something other than “being judged as M”, it seems like we cannot make a decision without substantive moral evaluation. Here we intuitively think that whether to accept or reject it is dependent on morally assessing whether x would be M with N or whether x would not be M without N. If quasi-realists are right that the mind-dependence counterfactuals (and their negation), as well as any other counterfactuals with a similar form (and their negation), express our moral commitments, then they should tell us why there is such an intuitive difference between them. Why is our intuition explicitly against the mind-dependence counterfactuals being moral sentences, but not other counterfactuals with a similar form?

It is not clear whether quasi-realists can explain our misinterpretation of it while avoiding intolerably ad hoc postulations. It seems that among all the counterfactuals with this form, it is just the mind-dependence counterfactuals, and only the mind-dependence counterfactuals, that tend to be explicitly mistaken by us as non-moral sentences. We can hardly find any general features of them, which can be shared by other counterfactuals with a similar form, that can explain our supposedly misleading intuition about them. It looks like the best we can have is an ad hoc explanation. Clearly, metaethical cognitivists can give us a far more plausible explanation about it. They can claim that most of us at least implicitly believe there is a realm of judgment-independent moral facts. The mind-independence claim is a metaethical claim derived from this metaethical belief about the nature of morality per se, and thus we can accept it without engaging in substantive moral assessment. They can also explain why the acceptance and rejection of other counterfactuals with a similar form require substantive moral evaluation: For those counterfactuals, we must investigate whether the moral property M actually supervenes upon the natural property N. If quasi-realists cannot give us at least an equally plausible explanatory story about this intuitive difference, we have good reason to reject the quasi-realist internal reading of the mind-independence claim.

4 The Conceptual Status

In this section, I consider an expressivist reply to the objection presented above. Expressivists may admit that we have the seemingly misleading intuition, but deny that this has any implication against the quasi-realist internal account. They can argue that the mind-independence claim, though being a moral claim, enjoys a conceptual status, in the sense that the acceptance of it is constitutive of engaging in moral discourse. Given such a conceptual status, competent moral talking presupposes the acceptance of this particular moral claim. This may explain why we have the intuition that people would not reject moral mind-independence no matter what radical changes have been made to their moral outlook. If this is the case, then there is no need to accuse them of misinterpreting a moral claim as a metaethical one.

However, how could a first-order moral claim enjoy a conceptual status? It seems that moral questions can only be answered by substantive moral theorizing rather than conceptual analysis, which means first-order moral truths cannot be conceptual truths. This is why Zangwill (1994) defends the conceptual status of moral mind-independence, but takes this privileged status as evidence against the quasi-realist internal reading. Expressivists need an argument in defense of a first-order moral claim being conceptual. There are cognitivists arguing for the possibility of knowing first-order moral truths by conceptual means (Cuneo & Shafer-Landau, 2014), but this cognitivist strategy is clearly not available to expressivists. Metaethical cognitivism allows the possibility of knowing an object of moral evaluation, under a certain description, is M by grasping the moral concept M because cognitivists believe that first-order moral commitments are just like ordinary descriptive beliefs.Footnote 7 However, if expressivism is true, then moral sentences are supposed to express conative attitudes. It is true that mastering moral concepts requires us to correctly believe what conative attitudes are supposed to be expressed by them, but it seems unlikely we can come to endorse certain conative attitudes simply by having these beliefs, given that conative attitudes and beliefs are distinct mental states.

Certainly, this just means if expressivism is true then we cannot come to endorse a moral commitment simply by reflecting on our moral concepts. Expressivists can still maintain that the endorsement of a particular moral commitment is a prerequisite for using moral concepts, that is, if we do not endorse it then we cannot be counted as using moral concepts at all. There is an argument in defense of this idea given by Sinclair (2008). Sinclair also agrees that the mind-independence claim should be interpreted as a moral claim, the function of which is to regulate the formation of moral commitments. However, he argues that there is a constitutive role of moral statements, that is, to mutually coordinate our attitudes and actions. Our moral concepts can play such a role only if we accept the moral commitments expressed by the mind-independence claim. It means if we reject moral mind-independence, then we are not able to use moral concepts to play their constitutive role. Therefore, the acceptance of the mind-independence thesis is constitutive of using moral concepts, and this is why moral mind-independence enjoys a conceptual status (pp. 273–276).

Nevertheless, I do not think this argument can successfully explain why we intuitively believe that the mind-independence claim has a privileged status. What Sinclair’s argument has defended is that the commitment to moral mind-independence is constitutive of using moral concepts, rather than knowing how to use moral concepts (it is difficult to see how the adoption of certain conative attitudes could be constitutive of knowing how to talk about morality). If Sinclair is right and a commitment to moral mind-independence is constitutive of using moral concepts, then we can rightly believe that the correct usage of moral concepts requires our moral commitments to be regulated by the mind-independence claim, thus know about how to use moral concepts, but reject moral mind-independence because we currently do not want to talk about morality.Footnote 8 This is just like we can know that attributing a racist slur requires us to have a pejorative attitude, but do not adopt this attitude because we do not want to attribute this slur to anyone. It follows that whether we should assume people are committed to the mind-independence claim is dependent on whether they actually intend to engage in moral evaluation and are prepared to form moral commitments. It is not necessary to attribute a commitment to moral mind-independence to them when they are not going to think and talk in moral terms.

However, our intuition about the privileged status of the mind-independence claim does not seem to covary with the belief about whether people are actually going to form any moral commitment. We can adjust the original scenario giving rise to the intuition slightly to make this clear. Once again, let us suppose that there is a group of people who accept the mind-independence claim as we do. One day a catastrophe radically reshaped their moral character. However, it is not that they have actually changed their moral views on everything, it is just that they have acquired completely different moral dispositions. By “different moral dispositions” I mean if they morally evaluate the same object as before, they will endorse a different moral view about it. However, they have no intention to morally reassess everything currently, and thus they have not actually formed any new moral commitment yet.

If Sinclair is right, then given that the mind-independence claim is a moral claim, we shall assume that people of this community now no longer have any disposition to negate the mind-dependence counterfactuals. Furthermore, since they currently do not intend to engage in moral inquiry and apply moral concepts, we do not have to assume that they have reacquired the commitment to moral mind-independence or the disposition to endorse moral mind-independence. However, this still sounds intuitively implausible. Would we be disposed to deny moral mind-independence simply because we have acquired a different moral character that gives us different moral dispositions? People with a vicious moral character may be inclined to make fundamentally different moral judgments from ours, but we all tend to (perhaps except quasi-realists) believe that they would also be disposed to acknowledge the mind-independence of morality as long as they take their moral views seriously. It is still quite counterintuitive to claim that one day we will be disposed to accept that what is morally right is counterfactually dependent on what we think to be right simply because of a transformation of moral character. All these suggest that we intuitively believe one can acquire utterly different moral dispositions while still preserving one’s disposition to endorse the mind-independence claim, regardless of whether one actually intends to morally reassess everything or whether one has actually endorsed any different moral views. If quasi-realists insist that the mind-independence claim is a moral claim, then they still owe us a satisfactory explanation of this allegedly misleading intuition. Arguing that the acceptance of moral mind-independence is constitutive of using moral concepts is of no help for them to meet this challenge because this intuition persists even when the moral appraiser is not going to apply moral concepts.

5 Conclusion

Quasi-realists attempt to accommodate the realist-seeming features of our moral discourse within the framework of expressivism. A major difficulty in achieving this goal is to give an expressivist-friendly interpretation of many ordinary thoughts of competent moral speakers that seem to have realist implications. Quasi-realists typically respond to this difficulty by interpreting many seemingly metaethical claims as first-order moral claims. In this article, I argue against the quasi-realist “internal” interpretation of the mind-independence claim about morality by demonstrating that quasi-realists cannot plausibly explain people’s tendency to treat moral mind-independence as a metaethical notion. A general implication following from my arguments is that the mere compatibility between the internal reading and the metaethical commitments of quasi-realists does not automatically justify the plausibility of this reading, especially in cases where our intuition strongly counts against the quasi-realist interpretation. What we need is not merely a metaethical theory that can allow us to say what we would ordinarily say about morality, but a metaethical theory that can correctly explain what we mean when we talk about morality. To defend their metaethical position, quasi-realists should convince us that their moral semantic theory can more accurately capture what we mean when making certain seemingly metaethical claims than the cognitivist alternative, and this is by no means an easy task.