Ratio. 2020;00:1–9. wileyonlinelibrary.com/journal/rati
|
11 | INTRODUCTION
Metaethical realists believe that moral facts and properties exist (Enoch, 2011; FitzPatrick, 2008; Scanlon, 2014; Shafer-Landau, 2003). Despite their agreement on this point, metaethical realists disagree about whether moral facts have ontological import. Two of the most prominent positions in this debate are those of the robust realist and the quietist realist.
Robust realists accept that moral facts have ontological import (Enoch, 2011; FitzPatrick, 2008; Jackson, 1998). What it means for facts, including moral facts, to have ontological import has been a point of contention in con-temporary analytic philosophy at least since the publication of W.V.O. Quine's seminal paper ‘On What There
O R I G I N A L A R T I C L E
Quietist metaethical realism and moral
determination
Wouter Floris Kalf
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2020 The Authors. Ratio published by John Wiley & Sons Ltd. Institute for Philosophy, Leiden University,
Nonnensteeg 1-3, Leiden, 2311 BE, the Netherlands
Correspondence
Wouter Floris Kalf, Institute for Philosophy, Leiden University, Nonnensteeg 1-3, 2311 BE Leiden, the Netherlands.
Abstract
Metaethical realists believe that moral facts exist, but they disagree among themselves about whether moral facts have ontological import. Robust realists think that they do. Quietist realists deny this. I argue that quietist realism faces a new objection; viz., the moral determination objec-tion. This is the objection that general moral facts (or moral principles) must determine specific moral facts (or which actions in the world are right and wrong) but that general moral facts cannot do this if they lack ontological import. I also argue that some robust realists can answer the moral determination objection. This gives these robust realists an edge over quietist realism.
K E Y W O R D S
moral determination, ontology, quietist realism, robust realism, T.M. Scanlon
Is’ (Quine, 1948). Robust realists tend to accept that an entity has ontological import if, and only if, it exists in the world. For example, William FitzPatrick writes that ‘certain elements of the world just are value-laden’ (FitzPatrick, 2008, p. 197). Within the camp of robust realism, we can distinguish between views that affirm that moral facts are irreducibly normative, such as Enoch's and FitzPatrick's, and views that deny this, such as Jackson's. Following Jonas Olson, I analyse irreducibly normative facts as follows (Olson, 2014, pp. 117–119). First, irreducibly normative facts are normative, which means that they give us a reason to act. For example, the fact that it is morally wrong to eat meat entails that there is a reason to refrain from eating meat. Second, irreducibly normative facts are irreducibly normative. This means that the reasons that are entailed by moral facts are not contingent on agents’ desires. Whether or not the agent desires to refrain from eating meat, they have a reason to do this if it is a moral fact that eating meat is morally wrong. In this paper, I am concerned with robust realists who say that moral facts exist in the world and that these facts are irreducibly normative.
Quietists typically also think that moral reasons are irreducibly normative (Dworkin, 1996; Parfit, 2012; Scanlon, 2014). Take T.M. Scanlon, on whose formulation of quietism I shall focus in this paper (henceforth, all references to page numbers only are to Scanlon's Being Realistic about Reasons). Like the robust realists, Scanlon says that the property of ϕ-ing being pro tanto morally wrong consists in there being a reason for an agent to re-frain from ϕ-ing (p. 19). He also says that ‘there are irreducibly normative truths about reasons’ (p. 51). But Scanlon thinks that when we say that moral reasons exist, ‘our ontological commitments … do not represent a claim on our part about what the world contains’ (p.24, emphasis in original). I will explain Scanlon's positive account of the way in which moral facts exist in the next section. For now, I just want to make the point that because Scanlon says that talking about moral reasons does not raise ontological questions about the world, he directly contradicts the robust realist's claim that moral reasons exist in the world. This is the sense in which quietists are quiet about the ontology of moral reasons. Moral reasons exist, but talking about them does not raise the same sort of ontological questions as talking about, say, elephants, which exist in the world.
In this paper, I argue that quietism faces a new objection, which I call the moral determination objection. To this end, I first say more about quietist realism (§2). I then introduce the moral determination objection, which is that general moral facts (or moral principles) must determine specific moral facts (or which actions in the world are right and wrong) but that general moral facts cannot do this if they lack ontological import (§3). In the same section, I argue that some robust realists have a satisfactory answer to the moral determination objection. In the penultimate section, I argue that three quietist responses to the moral determination objection fail (§4). I conclude that since quietists cannot and some robust realists can account for moral determination, we have a reason to prefer these robust realist views to the quietist alternative (§5).
2 | QUIETIST REALISM
Quietists accept that moral facts and properties exist but they deny that this raises the same sort of ontologi-cal questions that are raised if you accept that elephants exist in the world. But although quietists agree on this negative claim, they disagree on the positive claim of how to understand the mode of existence of moral facts. In this section, I describe Scanlon's account and I indicate how it differs from Ronald Dworkin's and Derek Parfit's versions of quietism.
Scanlon distinguishes a range of domains of human inquiry, including mathematics, science, and moral and practical reasoning (p. 19). It is natural to conceive of these domains as consisting of a realm of objects of a cer-tain kind and their properties (p. 19). If we do this, then we are committed to the claim that ‘all of our ontological commitments must be understood as claims about what exists in the physical world of space and time’ (p. 17). But we should not understand domains like this because this spells trouble for metaethical realists. The trouble is that if there were moral facts, ‘then they would be entities or qualities or relations of a very strange sort, utterly different from anything else in the universe’ and so they would be ‘too queer to be instantiated’
(Mackie, 1977, p. 38; Olson, 2014, p. 12n17). Therefore, says Scanlon, ‘a domain is better understood in terms of the kind of claims it involves, and hence in terms of concepts that it deals with, such as number, set, physical object, reason, or morally right action’ (p. 19).
Yet for Scanlon, talking about domains does not only commit you to a certain semantics. We may, and should, also do metaphysics. Scanlon accepts Quine's account of metaphysics according to which your metaphysics give you a theory about what exists (this is your ontology) as well as a theory about which concepts you can express within that theory (this is your ideology) (Quine, 1951; Scanlon, 2014, p. 30). According to Scanlon, each of the various domains raises its own ‘metaphysical and ontological questions’ (p. 25). For example, the domain of natural science has implications for the physical world and thus for our ontology. Contrastingly, the moral domain has ‘external presuppositions of a different kind’ (p. 22). This is because the moral domain presupposes that there are moral standards that everyone has good reason to take seriously (p. 22). Whether this presupposition is correct is determined by the domain of practical reason, and both the domain of morality and the domain of practical reason feature in our metaphysics as part of our ideology, not as part of our ontology (p. 30). So, pace Dworkin, we can ask external questions about the various domains (Dworkin, 1996). It is just that in the case of the moral domain, the only external questions that are raised concern the domain of practical reason and thus our ideology. The moral domain does not raise external questions about the physical domain, and so it does not raise ontological questions about that domain.
The fact that the various domains are connected through this relation of external presupposition is the ex-ception, not the rule. Apart from the external presupposition relation that may connect one domain (e.g., that of morality) to another (e.g., that of practical reason), each domain is fully autonomous in the sense it sets its own standards for good reasoning within the domain. Thus, the domain of mathematics determines what good math-ematical reasoning looks like, and the domain of morality determines what good moral reasoning looks like. As a corollary, Scanlon argues that questions about whether certain entities exist are settled by standards within the domain in question:
The claim that mountains exist is licensed by and licences certain other claims about the physical world. The claim that there exists a number … is licensed by and licences certain other mathematical claims. And in each case that is all there is to it. (p. 25)
The same holds for the domains of practical reason and morality. For Scanlon, standards ‘consist of substantive principles about the domain’ and are ‘justified by less explicitly codified reasoning about the subject matter in ques-tion’ (p. 20). The standards are also revisable ‘in the light of further reasoning of its kind’ (p. 20). Nevertheless, and pace Parfit, all of this does not entail that reasons and obligations ‘exist in a different sense’ than elephants (Parfit, 2012, Vol. 2, p. 469). As such, Scanlon avoids a possible objection that Parfit faces, which is that showing that reasons exist in a different sense than the sense in which elephants exist is a Pyrrhic victory.
Scanlon further complicates this picture, and rightly so I think, by distinguishing between pure and mixed statements in the various domains. This is a complication because the thesis that the truth values of statements in a certain domain are settled exclusively by standards internal to that domain only applies to pure statements. An example of a pure statement in the moral domain is ‘what it takes for an action to be morally wrong is that the action has certain consequences.’ An example of a pure statement in the domain of practical reason is ‘irreducibly normative reasons exist.’ An example of a mixed statement is ‘this action in the world is morally wrong.’ This last statement involves not only the claim that an action having certain consequences would be wrong (settled by standards internal to the moral domain) but also that this action actually has those consequences (settled by the standards for good empirical observations in the physical domain) (p. 21). Thus, whereas the truth value of a pure statement is determined wholly by the standards internal to that domain, the truth value of a mixed statement is determined by the relevant standards of the appropriate domains working together.
We can now summarise the essential commitments of Scanlon's view (Wodak, 2017, pp. 2795–2798). Scanlon is committed to:
Diversity: Different statements have different subject matters, and what it takes for a statement to be true
depends on its subject matter.
Autonomy: There are no privileged first- or second-order domains.
Consistency: If domains α and ß contain, entail, or presuppose inconsistent statements—if p is true in α but
false in ß—either α or ß must be rejected.
Scanlon accepts Diversity because he thinks that the claim that elephants exist is licensed by claims about the physical world, that the claim that numbers exist is licensed by mathematical claims, and so on (p. 25). Scanlon ac-cepts Autonomy because Autonomy blocks an important route to scepticism. If Autonomy holds, then the sceptic cannot prioritize the standards of the physical domain that determine whether something exists in that domain and use these standards to cast doubt on the claim that numbers and reasons exist (p. 19). The various domains that we have talked about so far (morality, mathematics) are first-order domains. An example of a second-order domain would be that of ‘existence’ or ‘the universe’ at large. Scanlon rejects the idea that there are such second-order domains because he thinks that ‘genuine ontological questions are all domain-specific’ (p. 24n10). So really, I think, Autonomy says that there are no privileged domains, where the only domains that exist are first-order domains.
Finally, Scanlon accepts Consistency because without Consistency, we would be able to dream up a large number of diverse and autonomous domains, such as the supernatural domain that would license us to say that witches exist (p. 21). But this supernatural domain stands in the external presupposition relation to the physical domain because the supernatural domain licenses ‘claims about events in the physical world and their causes’ (p. 21). And since the claims about the physical world that are presupposed by the claims in the supernatural domain ‘conflict with claims of physics and other empirical sciences’ we have ‘decisive reason to reject the idea that there are witches’ (p. 21).
It is not clear whether Scanlon may reject the supernatural domain by prioritizing the domain of science given that he accepts Autonomy and Consistency (Wodak, 2017, pp. 2806–2807). If Consistency and Autonomy are doing so much important work in Scanlon's argument, then we might as well reject claims about science in order to keep our preferred claims about witches and their supernatural powers. Nevertheless, this is not my objection and I will assume that Scanlon can solve this issue. My objection is the moral determination objection, which I detail below.
3 | THE MOR AL DETERMINATION OBJECTION
According to the moral determination objection, Scanlon must explain how general moral facts (or moral princi-ples) determine specific moral facts (or which actions in the world are right and wrong) but general moral facts cannot do this if they lack ontological import. The judgment that stealing this apple in the world is wrong is a mixed judgment, so we need to know the facts in the physical universe that entail that this act has the properties that would make it wrong, such as having certain consequences. I am going to give Scanlon whatever he needs with respect to these physical aspects of the mixed judgment. Instead, I worry about how general moral facts can determine the specific moral fact that stealing the apple in this world is wrong.
Why think that quietists like Scanlon must explain how general moral facts determine specific moral facts in the world? The answer is that this follows from the background picture in metaethics that all realists accept. To see this, first consider the alternative background picture in metaethics that anti-realist, constructivists accept. Sharon Street argues that the explanation for why the judgment that you must not perform action A (morally speaking, or perhaps all-things-considered) is correct is that it withstands scrutiny from the practical point of view
(Street, 2008). As such, there is no need to say that there are moral facts at all. Contrastingly, as realists, quietists think that the explanation for why the judgment that you must not perform action A (morally speaking, or perhaps all-things-considered) is true is that it is a moral fact that performing that action is morally wrong and that true judgments correspond to these facts. And like many other realists, quietists do not end there, accepting the claim that it is a brute fact that action A is morally wrong. Quietists explain the wrongness of action A, as it occurs in the world, by reference to the general moral fact that this action has certain consequences (or whichever general moral fact they may accept). And this moral fact, according to the quietist, exists because saying that it exists is licenced by the standards internal to the moral domain. So, quietists must explain how general moral facts deter-mine specific moral facts in the world.
But why think that quietists like Scanlon must explain how general facts determine specific moral facts in
the world? Why can't Scanlon insist that general moral facts determine specific moral facts and that all of this is
internal to the moral domain? The answer here, as I explain below, is that if quietists keep all of this internal to the moral domain, then their view collapses into quasi-realism.
Why think that quietists like Scanlon cannot explain how general moral facts determine moral facts or which actions in the world are right and wrong? Basically, because they place general moral facts or moral principles like ‘what it takes for an action to be morally wrong is that the action has certain consequences’ in the moral domain and because they sever all relations between that domain and the physical domain, thus making it impossible for the moral determination relation to hold between general moral facts in the moral domain and the actions in the physical world that are supposed to be wrong. It is all well and good that Scanlon has an account of how the moral standards internal to the moral domain determine which actions, if they obtain, are right and which ones are wrong. I explained this account in the previous section. Scanlon can explain how moral standards explain which specific moral facts obtain, and all of this stuff is internal to the moral domain. This is what I mean when I say that Scanlon can explain how moral standards determine which moral actions ‘if they obtain’ are right and which ones are wrong. If we remain in the moral domain, then Scanlon can explain how general moral facts determine which actions would be wrong if they obtained in the world. But Scanlon should also have an account of how the moral standards internal to the moral domain determine which actions, as they happen in the world, are right and which ones are wrong. I think that Scanlon cannot explain this.
Schematically, my worry is that Scanlon should but cannot explain how GMF: all actions with consequences c in contexts c’ are wrong determines
SMF: stealing this apple in the world is wrong.
In this scheme, GMF is the general moral fact or moral standard that determines specific moral facts, such as the specific moral fact SMF that stealing this apple in the world is wrong. I think that this moral determination relation should be a metaphysical relation. After all, Scanlon, as a realist, is doing metaphysics. But Scanlon cannot say that the moral determination relation is a metaphysical relation because he severs the domain of morality from the physical domain. The only type of relation that Scanlon can appeal to is a non-metaphysical, first-order moral relation, but if he does then, then his view collapses into a version of quasi-realism. But what is a metaphysical relation? We cannot answer this question in a context-independent way. In some metaphysical debates we want to answer this question in one way but in some other debates we are going to want a different answer. I therefore propose that we focus on the context of our discussion, which involves metaethical realists and metaethical qui-etists. At this point, a comparison with the robust realists will be useful.
Robust realists like David Enoch also need an account of moral determination. As Enoch puts it: It would be wrong of you not to give that grad student feedback on her paper in a timely manner. Why? Well, because she legitimately expects this feedback, and not getting it on time will harm her
progress. But this is not all—more is needed. What makes it the case that it would be wrong of you not to give that grad student her feedback soon is all of that, together with the fact that you ought not to harm people by frustrating their legitimate expectations. (Enoch, 2019, p. 1)
In other words, the very general moral fact or moral principle that you ought not to harm people by frustrating their legitimate expectations determines the specific moral fact that you ought to give your grad student feedback on time (together with some physical facts, but as I said at the start of this section, I will assume that Scanlon can have whatever he needs here). The question for Enoch is whether and if so, how he can explain this determination relation.
According to Enoch, his version of non-naturalism can explain how moral principles ground specific moral facts by distinguishing between moral and metaphysical grounding, by explaining that only some contexts require a metaphysical grounding, and that the analogy with legal grounding in the metaphysical sense tells us all we need to know about moral grounding in the metaphysical sense (Enoch, 2019, pp. 6–12). In the example cited above involv-ing timely feedback to the graduate student, all we need for moral groundinvolv-ing is the claim that you'd be wronginvolv-ing the graduate student if you do not give her feedback on time in virtue of the fact that not doing that would harm her progress—no mention of the general moral principle here. But a full metaphysical explanation would contain a reference to this moral principle and how it grounds specific moral facts.
This contrasts with a first-order account of the moral determination relation. First-order moral commitments express your views on moral questions and do not do metaphysical work. Some quietists think that they should use this idea and expel all metaphysics from their metaethical theory. For instance, Dworkin's version of quietism entails that any ‘philosophical-sounding proposition that there are moral properties’ is actually an ‘I-proposition’, where an I-proposition expresses a ‘positive moral judgment’ (Dworkin, 1996, p. 100, p. 92). But neither Dworkin nor Scanlon should accept this, for then their view collapses into quasi-realism, as I will now argue.
Quietism collapses into quasi-realism if it accepts that all seemingly metaphysical statements are first-order moral statements in disguise. The quasi-realist ‘tries to earn, on the slender basis (of an anti-realist metaphysics), the features of moral language … which tempt people to realism’ (Blackburn, 1984, p. 171). According to Simon Blackburn, we can
know moral propositions to be true. Or, really true, or really factually true, or really in accord with the eternal harmonies and verities that govern the universe (Blackburn, 1998, p. 79).
Thus, quasi-realists take this to the extreme by denying that any question that looks like a metaphysical question is in fact a metaphysical question. For quasi-realists, there is not a grain of metaphysics in their position, neither in the form of ontology nor in the form of ideology. Quasi-realists just have a fancy semantics that tells them that seemingly metaphysical judgements are actually first-order moral judgments.
This contrasts with quietists, who do have some metaphysics in their theory. If quietists insist that their view is not just semantics but also metaphysics—as they should, for then their view does not collapse into quasi-re-alism—but if they do not also give moral facts a metaphysical job, such as the job to explain how general moral facts determine which specific acts in the world are wrong and which ones are right, then quietists are pushing a distinction without a difference. And since distinctions without a difference do not count, it follows that quiet-ism collapses into quasi-realquiet-ism if they do not account for the moral determination relation. This is another way of putting my point that as realists, quietists should explain moral determination. If quietists do not explain moral determination, then their view collapses into quasi-realism.
I am now in a position to state the moral determination objection in full. The moral determination objection takes the form of an explanatory challenge for quietists: explain how general moral facts determine specific moral facts but without appealing to just our moral attitudes, as that would cause their view to collapse into quasi-real-ism. As such, the problem for quietists is twofold. The first part of their problem is that, unlike the robust realists, they do not place moral properties in our world of space and time. This means that they cannot appeal to worldly
metaphysical relations to explain moral determination. The other part of their problem is that quietists only postu-late the external presupposition relation between the various domains. Even if we think that this is a metaphysical relation, then we must still say that this is not even a candidate relation that might explain moral determination, for a relation that checks whether the commitments in one domain jibe with those in another does not even look like a relation that could determine right and wrong.
Some robust realists can explain the moral determination relation and that this therefore gives these robust realists an edge over the quietist realist. Above I already indicated how Enoch can account for the moral deter-mination relation. Below I explain how Jonathan Dancy, a non-naturalist realist particularist who does not accept moral principles, can explain moral determination. I take this to show that moral determination is a general phe-nomenon that different realists will try to account for within their own theoretical framework: Enoch appeals to moral principles, Scanlon appeals to general moral facts or moral standards, and Dancy appeals not to moral principles but to the notion of resultance.
Dancy may use the notion of resultance to claim that the moral properties of an action exist ‘in virtue of … some other’ properties, such that these other properties are the resultance base for the moral properties (Dancy, 1993, p. 73). When Dancy talks about resultance, he talks about the reason-making relation between moral and non-moral properties in the world. Crucially, resultance is different from supervenience, for the latter is a relation of
modal co-variance and not, like resultance, a metaphysical relation that explains what exists in virtue of what. As
Jussi Suikkanen puts it, resultance is a ‘metaphysically robust, worldly making-relation’ (Suikkanen, 2010, p. 102). Robust realists can appeal to resultance to explain moral determination. What makes an action wrong, in a meta-physical sense of making something the case, is that the action would instantiate or bring about a moral property ‘that results from some of its non-moral properties’ (Dancy, 2000, p. 132). The moral fact that results from the natural facts determines that it is wrong to perform any action that brings about this constellation of natural facts. Since all of this happens in the world of space and time, we have a metaphysical moral determination relation.
Note that Dancy's and Enoch's positions may fail for reasons that have little or nothing to do with their ability to explain moral determination. However, in the context of this paper, all I need is the claim that theories like theirs are at least minimally plausible, which they certainly are, and that these views can explain moral determination. If quietism is also minimally plausible, which it certainly is, but if quietism cannot explain moral determination, then all else being equal, robust realism has an edge over quietism. Finally, note that not all robust realist views may be able to explain moral determination. These robust realist views do not gain relative traction from the quietist's failure to explain moral determination.
4 | QUIETIST RESPONSES TO THE MOR AL
DETERMINATION OBJECTION
In this section, I consider three responses to the moral determination objection. The first is that Scanlon can sim-ply postulate the moral determination relation as an extra relation between the moral and the physical domain, in addition to the external presupposition relation that he postulates between the moral domain and the domain of practical reason. Given that Scanlon already has the external presupposition relation, there is no principled barrier to the postulation of such an additional relation. The idea would be that the moral facts as they exist in the moral domain somehow determine that actions in the world are wrong. The all-important question though is what this relation might look like.
Quietists are free to conceive of the moral determination relation from within their own theoretical frame-work. But at least the moral determination relation must have objects or properties as its relata, just as the re-sultance relation as used by some robust realists has moral and non-moral properties as its relata. Saying that the moral determination relation must have objects or properties as its relata assumes that domains are realm of objects and their properties. But as we saw in §2, quietists should not say this, for that invites Mackie's worries
about robust realism. Instead, we should understand a domain ‘in terms of the kind of claims it involves, and hence in terms of concepts that it deals with, such as number, set, physical object, reason, or morally right action’ (p. 19). The problem with accepting this account of domains is that we cannot postulate relations between the domains that have objects or properties as their relata. We would be back to using something like the external presupposi-tion relapresupposi-tion, but that relapresupposi-tion clearly cannot explain moral determinapresupposi-tion.
At this point, Scanlon might push a second response. He might say that I am conceiving of this theory through the lens of the robust realist and that this is dialectically unfair. I borrow the robust realist's conception of the moral determination relation that demands that it has objects and properties as its relata and then I demand that quietists give us that. Obviously, quietists cannot do this. Moreover, Scanlon might say that what quietists do offer is enough. The full quietist package is a successful metaethics, perhaps because it explains the full form and shape of ordinary moral discourse, as Jackson insists, or perhaps because, as Scanlon puts it, it accounts for the desid-erata for a plausible metaethics that ‘those we talk with most frequently’ accept (Jackson, 1998; Scanlon, 2014, p. 2). But whatever we think determines the desiderata for a successful metaethics, Scanlon's response to the moral determination objection cannot consist in simply denying that he has to say anything at all about moral determina-tion. For I have already argued that quietist realists need to account for moral determination (§3).
Third and finally, quietists might argue that they can treat all seemingly metaphysical claims like ‘it is a fact that stealing is wrong’ as a fancy way of formulating purely moral claims like ‘this act is wrong’ and avoid collapsing into quasi-realism. Quietists might want to say this because this would also allow them to ignore the moral determina-tion objecdetermina-tion. After all, if seemingly metaphysical claims really are moral claims in disguise, then quietists do not have to show how these claims do metaphysical work, and a forteriori they do not have to show how these claims explain moral determination. How can quietists say this and avoid collapsing into quasi-realism? One suggestion is that quietists might turn to the cognitivism-expressivism debate (Dworkin, 1996, pp. 108–112; Scanlon, 2014, pp. 53–68). They might consider the claim that ‘realism is a form of cognitivism’ and that ‘every realist is a cogni-tivist’ (Shafer-Landau, 2003, p. 17). As quietists are cognitivists and as quasi-realists are expressivists, we have the distinction between quietism and quasi-realism that we are after, and without mentioning moral determination.
But this does not work. Contrary to what Shafer-Landau thinks, there are many interesting debates in meta-ethics that presuppose that you can pick and choose your preferred semantics on the basis of either a realist or an anti-realist metaphysics. Consider the case from the point of view of anti-realist metaphysics first. There is now a growing literature on the question which semantic theory should be accepted by moral error theorists, who are cognitivist anti-realists and believe that all our first-order moral judgments are false (Joyce, 2001; Kalf, 2018; Olson, 2014). This so-called ‘now what question’ is important for moral error theorists because if they stick to their semantic cognitivism on the basis of their metaphysical anti-realism, then they carry ‘on believing in some-thing the evidence of whose falsity they have seen and accepted,’ which is ‘irrational’ (Joyce, 2001, p. 178). One option for committed moral error theorists is to stop expressing false beliefs and to start expressing conative attitudes, which is what ‘revolutionary expressivists’ recommend (Svoboda, 2017, p. 48). Another option for moral error theorists is to fictionalize their moral thoughts and utterances, thereby pretending as if there are moral truths rather than continuing to communicate false beliefs (Joyce, 2001, pp. 206–231). Revolutionary expressivists and revolutionary fictionalists add a non-cognitivist semantics to an anti-realist metaphysics where, as error theorists, they used to accept a cognitivist, truth-apt semantics. Thus, this entire debate is premised on the assumption that we can pick and choose our moral semantics on the back of our prior commitments about moral metaphysics (Kalf, 2019).
What is more, and moving beyond error theory, it has even been argued that robust realists, who are meta-physical realists, should consider changing their moral semantics if doing this maximizes the amount of good that moral agents can procure (Ingram, 2015). If robust realists truly care about the good, and if they learn that one is more likely to do good if one stops expressing beliefs with moral judgments, then one should, morally speaking, stop expressing beliefs with moral judgments. If that's right, then in these circumstances, robust realists in meta-physics should become expressivist in semantics. This too seems to be coherent and philosophically interesting.
So, it is not true that all realists are cognitivists or that all quasi-realists are expressivists. And so, we cannot dis-tinguish the quasi-realist from the realist on the basis of their preferred semantics. Therefore, the third and final responses to the moral determination objection fails.
5 | CONCLUSION
Quietist realists face the moral determination objection. They should, but cannot, answer the objection that gen-eral moral facts (or moral principles) must determine specific moral facts (or which actions in the world are right and wrong) but that general moral facts cannot do this if they lack ontological import. Robust realists face the same challenge, but some of them can account for moral determination. This gives these versions of robust realism an edge over quietist realism.
ORCID
Wouter Floris Kalf https://orcid.org/0000-0001-8636-3264 REFERENCES
Blackburn, S. (1984). Spreading the word. Oxford University Press. Blackburn, S. (1998). Ruling passions. Oxford University Press. Dancy, J. (1993). Moral reasons. Blackwell.
Dancy, J. (2000). The particularist's progress. In B. Hooker & O. M. Little (Eds.), Moral particularism. Clarendon Press. Dworkin, R. (1996). Truth and objectivity: You'd better believe it. Philosophy and Public Affairs, 25, 87–139.
Enoch, D. (2011). Taking morality seriously. Oxford University Press.
Enoch, D. (2019). How principles ground (Oxford Studies in Metaethics, Vol. 14). Oxford University Press.
FitzPatrick, W. J. (2008). Robust ethical realism, non-naturalism, and normativity (Oxford Studies in Metaethics, Vol. 3). Oxford University Press.
Ingram, S. (2015). After Moral Error Theory, After Moral Realism. The Southern Journal of Philosophy, 53, 227–248. Jackson, F. (1998). From metaphysics to ethics. Oxford University Press.
Joyce, R. (2001). The Myth of Morality. Cambridge University Press. Kalf, W. F. (2018). Moral error theory. Palgrave MacMillan.
Kalf, W. F. (2019). The belief problem for moral error theory. Inquiry (online first). Mackie, J. L. (1977). Ethics. Penguin Books.
Olson, J. (2014). Moral error theory. Oxford University Press. Parfit, D. (2012). On what matters (Vol. 2). Oxford University Press. Quine, W. V. (1948). On what there is. The Review of Metaphysics, 2, 21–38.
Quine, W. V. O. (1951). Ontology and ideology. Philosophical Studies, 2, 11–15. https://doi.org/10.1007/BF021 98233 Scanlon, T. M. (2014). Being realistic about reasons. Oxford University Press.
Shafer-Landau, R. (2003). Moral realism: A defence. Oxford University Press.
Street, S. (2008). Constructivism about reasons (Oxford Studies in Metaethics, Vol. 8). Oxford University Press.
Suikkanen, J. (2010). Non-naturalism: The Jackson challenge (Oxford Studies in Metaethics, Vol. 5). Oxford University Press.
Svoboda, T. (2017). Why Moral Error Theorists Should Become Revisionary Moral Expressivists. Journal of Moral Philosophy, 14, 48–72.
Wodak, D. (2017). Why realists must reject normative quietism. Philosophical Studies, 174, 2795–2817. https://doi. org/10.1007/s1109 8-016-0811-3
How to cite this article: Kalf WF. Quietist metaethical realism and moral determination. Ratio. 2020;00:
