We frequently find that, to defend or to refute metaethical theories, we rely on the truth of other, independent arguments concerning central features of our moral life. The two most well-known examples are the use of arguments concerning moral motivation and the Humean theory of motivation to derail cognitivist moral realism.
Much ink has been spilled over whether these views are independently true, let alone whether, if true, they present genuine problems for moral realism. In this paper, I consider whether an overlooked feature of our moral life might provide a route to vindicating naturalist moral realism in much the same way that the aforementioned claims are often used to undermine it. The overlooked feature that I have in mind is that of moral status, which has received a great deal of attention in normative ethics and in some corners of metaethics. This seems promising, I believe, because moral status is a central feature of our moral lives, like (moral) motivation, and yet it does not require that we have any independent commitments in broader debates between moral realists and antirealists. What's more, there are influential, albeit incomplete, uses of the concept of moral standing for the very same metaethical purposes, which highlight its potential vindicatory import. Namely, Rawls's (Reference Rawls1980) use of the concept of the free person to justify constructivism as generating warranted, objective moral claims and aspects of Korsgaard's (Reference Korsgaard1996: 149–66; Reference Korsgaard2009: especially 204) constitutivism.Footnote 1 In a similar vein, Copp (Reference Copp, Beauchamp and Frey2011: 282) argues that that there are some basic claims about moral status—in particular, claims about animals’ moral standing—that our metaethical views should explain. That these important accounts, right or wrong, use the concept of moral status in this way suggests that we should investigate whether we can build a full-blown vindicatory argument around the concept of moral status in order to adjudicate competing metaethical views. It is this task that occupies me in this paper.
I take this to be an interesting strategy because, contrary to appearances, most metaethical views (should) grant that there is some property called moral status. Why? Metaethical views are concerned, broadly speaking, with the question of the possibility and nature of moral facts. To be concerned with this question, they must first grant that there are (or could be) creatures to whom we have moral obligations, which, in turn, may be underwritten by the very moral facts about which they are primarily concerned. In this regard, what they actually disagree about is whether and how moral status and the claims about our moral obligations to those that possess it are metaphysically underwritten. So, perhaps surprisingly, moral status (should) matter(s) to both moral realists and antirealists. If this is right, then it would be a point in favor of whichever view can more plausibly explain it.
But there are two trouble spots for this approach. First, we need a view of moral status that is ecumenical with regard to the metaethical debate—that is, it cannot depend on the idea that, say, there are mind-independent moral facts about moral status. Naturally, this approach would be circular. But, as I will argue, an ecumenical account can be had. The more interesting, second source of trouble arises when trying to square the vindicatory approach outlined above with the persistence of normative ethical disagreement about what constitutes moral status. Even a cursory glance at the literature on moral status reveals significant debate over what constitutes moral status, the characteristics that are necessary and sufficient for it, and who or what has it. And nothing about using moral status as a weighting consideration in the metaethical debate does or should resolve these debates. But the unresolved normative ethical disagreement suggests an additional issue with which we must contend in using this vindicatory strategy. Namely, if a metaethical view can explain the possibility of moral status, its explanation should be consistent with the persistence of this first-order normative ethical disagreement.
With these issues in mind, the plan for the paper is as follows. In section 1, I provide what I call an Ecumenical View of Moral Status. I show how it can serve to adjudicate metaethical disagreement in a way that parallels the use of the Humean theory of motivation and judgment internalism in these same debates. In section 2, I show why, after some significant modifications, this is a promising route for vindicating naturalist moral realism. In section 3, I compare naturalist moral realism's ability to account for moral status with quasi-realism's ability to do so, concluding that quasi-realism founders in its ability to explain persistent normative ethical disagreement about what constitutes moral status. This argument has three interesting results: first, we have a new route for metaethical vindication; second, we can vindicate naturalist moral realism in a fashion analogous to the use of the Humean theory of motivation and theories of moral motivation for vindicating noncognitivist antirealism; and third, quasi-realism's plausibility is undermined by an inability to explain disagreement, but not for the familiar reasons.
1. Moral Status's Metaethical Import
Broadly speaking, moral status describes the standing a creature has such that we owe it certain kinds of treatment.Footnote 2 So, to use a simplified version of Darwall's (Reference Darwall2006: 13) example, I ought not to stomp on my friend's foot in part because she has the moral standing to deserve not to be unfairly harmed, and stomping on her foot is unfairly harming her. The relevant question, however, is what grounds this standing. In this regard, theories of moral status are largely concerned with the criteria by which a creature gains this standing.Footnote 3 There are at least three types of (often competing) criteria for moral status. One view emphasizes a creature's advanced rational and other cognitive capacities, which underwrite her status as a rights-bearer. A second view emphasizes a creature's capacity to suffer and, thereby, to have the kind of life that possesses or lacks well-being. A third competing view emphasizes a creature's capacity to stand in care-giving relations with others (Jaworska Reference Jaworska2007; Jaworska and Tannenbaum Reference Jaworska and Tannenbaum2015). This debate is best represented, albeit only approximately,Footnote 4 by the disagreements among deontologists, utilitarians, and care ethicists. We typically think it is best answered by the normative ethical theories that we defend, although there have been notable stand-alone accounts of moral status (e.g., Darwall Reference Darwall2006; O'Neill Reference O'Neill1989).
Regardless of how this debate is resolved, I suggest that there is an analytically distinct set of metaethical claims that underwrite nearly all views about moral status. This does not mean that all views of moral status explicitly assent to these claims; rather, they highlight the metaethical import of moral status to which all such accounts must assent to be coherent. I call this set of claims the Ecumenical View of Moral Status. Below, I consider what these claims are and why they should, contrary to the current state of the literature, play a central role in evaluating metaethical theories.
1.1. An Ecumenical View of Moral Status
For a view of moral status to be ecumenical, it cannot make any contentful normative ethical commitments—say, about the characteristics that grant one moral status. It also cannot make any contentful metaethical commitments. Here it may be harder to see in what such a commitment might consist. Broadly speaking, for a view about moral status to make a contentful metaethical commitment would be for it to make an explicit commitment to the kind of metaethical view that would be best able to vindicate its truth.
In this regard, an Ecumenical View of Moral Status makes the following central claim: a creature possesses or is granted moral status by virtue of some set of characteristics she does or could possess, and this status is a fact about her about which we can have knowledge. To be complete, however, any view of moral status must assent to two additional claims, which highlight its metaethical import. The first, which I call The Proposition that Moral Status is Mind-Independent,Footnote 5 states that moral status is generated from a set of necessary and sufficient fact(s) about the creatures who possess them and thus that moral status is at least partially independent of our beliefs and judgments about the creatures in question.Footnote 6
Contrary to appearances, this claim does not require that we stake out a clearly moral realist position. Rather, the claim to mind-independence in this context only concerns the descriptive facts that underwrite moral status rather than their moral significance. One could endorse a subjectivist ideal advisor view, for example, and meet the requirements of this proposition. It is also consistent with a variety of noncognitivist antirealist positions: consider here quasi-realists, who take some set of mind-independent properties to generate projective, mind-dependent moral attitudes. Similarly, this claim is consistent with normative ethical theories that take moral status to be relational and thus in part dependent on others’ affective states. This is because, for these views, the relevant facts that ground moral status are mind-independent: namely, that we have these capacities or the relevant affective states is, itself, a mind-independent fact. It is only the content of these states that is mind-dependent.
They must also assent to a second claim—namely, that these descriptive, grounding facts are natural facts. Call this claim The Proposition that Moral Status is Grounded in Natural Facts. Note here that ‘natural’ need not be taken to mean that moral status is reducible to a set of natural, nonmoral facts or that moral facts are themselves an additional kind of natural facts. This claim simply states that moral status is fixed by a set of facts that are, broadly speaking, empirically accessible. So they may be facts about the natural world, our moral sentiments, the nature of emotive utterances, among others. In this regard, this proposition is also consistent with a variety of very different metaethical views.
But even if these two propositions are metaethically ecumenical, the reader may worry that they are not normative ethically neutral. There are normative ethical views about moral status that do not appear to assent to both or either of these claims. Here consider views that, like Darwall's (Reference Darwall2006), take moral status to derive from one's status as the (proper) addressee of moral claims. Similarly, strongly contractualist and contractarian views often—although not universally—take one's status as a contractor to guarantee one's moral status (Darwall Reference Darwall2006; O'Neill Reference O'Neill1989; Reference O'Neill2003; Rawls Reference Rawls1999; Scanlon Reference Scanlon1998). Yet, in the case of views like Darwall's, they tend to make a set of claims about the sophisticated reason-giving capacities of others who possess moral status, thereby placing this type of view closer to the two propositions above. The same can be said of contractualist and contractarian views, notwithstanding their largely constructivist orientation. Even though these views are not what both Shafer-Landau (Reference Shafer-Landau2003) and Street (Reference Street2006), respectively, call ‘stance-independen[t]’, nothing about a normative ethical theory that is underwritten by metaethical constructivism requires that we deny either or both of the two propositions above. This is because the characteristics that constitute moral status can be understood in a morally neutral, mind-independent wayFootnote 7 while granting that they gain moral significance only by virtue of their role in a procedure for determining our moral obligations. On both contractualist and contractarian views, the characteristics that underwrite our status as contractors can be understood in just this way.
That most views of moral status do (or at least could) agree on the truth of these two propositions is not, for my purposes, important for settling the debate about what constitutes moral status. In what follows, I show that these views’ consensus around these two claims is a useful outer constraint for the moral realist.
1.2. Moral Status and Metaethical Vindication (for the Moral Realist): An Attempt and a Correction
So how might we understand the relationship between a broad moral status account, as comprising these two propositions, and second-order metaethical theories? I propose examining how other, structurally cognate accounts of features of our moral lives are used in metaethical debates. Doing so provides a model for using the Ecumenical View of Moral Status and its attendant propositions to vindicate naturalist moral realism.
The most familiar place to find this type of argument is in the use of the Humean theory of motivation and (typically) judgment internalism either to vindicate noncognitivist forms of antirealism or to refute cognitivist moral realism. Though examples of the former and the latter (e.g., Blackburn Reference Blackburn1984; Hare Reference Hare1952; Mackie Reference Mackie1977) are too numerous to discuss in detail here, we can treat Russ Shafer-Landau's (Reference Shafer-Landau2003) brief sketch of this type of argument, which he calls the ‘Non-cognitivist Argument’, as a model:
1. Necessarily, if one sincerely judges an action right, then one is motivated to some extent to act in accordance with that judgment. (Motivational Judgment Internalism)
2. When taken by themselves, beliefs neither motivate nor generate any motivationally efficacious states. (Motivational Humeanism)
3. Therefore moral judgments are not beliefs. (Moral Non-cognitivism). (121; emphasis in the original)
If this argument is right, Shafer-Landau goes on to claim, it entails that ‘moral realism (and moral cognitivism generally) is false’ (121). Although it is certainly not true that all or even most metaethical views make exactly this argument, it is nonetheless the case that the Humean theory of motivation and/or views about moral motivation (both judgment internalism and externalism) figure nontrivially in metaethical debates about moral realism and antirealism.
Given how these views are used, we might think that the Ecumenical View of Moral Status and its attendant claims can serve a similar function for the moral realist.Footnote 8 Before considering this possibility, however, there are two objections to address. One might doubt that moral status, however ecumenically it is understood, can serve a similar function as the Humean theory of motivation. This is because the former is a normative claim whereas the latter is descriptive. Notwithstanding the justified worry about the naturalistic fallacy here, I contend that the Ecumenical View of Moral Status is descriptive rather than normative. To see why, consider three ways that the parallel with the Humean theory of motivation is appropriate. First, although specific accounts of which properties constitute moral status are normative ethical theories, the terms of the Ecumenical View of Moral Status makes no substantive normative ethical commitments of this type. Second, as I note earlier, the reason that moral status may (surprisingly) bear on metaethical debates is that moral status is a feature of the moral landscape on which (most) metaethical views, realist and antirealist alike, must take a stand. In this way, there is a natural parallel with the use of the Humean (or other) theor(ies) of motivation and moral motivation. Finally, the question of moral status is, at least for my purposes, a version of the more general metaphysical question concerning moral facts or properties. To use some version of the Ecumenical View of Moral Status to vindicate metaethical views is structurally analogous to using descriptive claims about the metaphysical nature of motivation to disprove cognitivist moral realism.
The second objection concerns whether using the Ecumenical View of Moral Status to vindicate metaethical views requires that we take a stand in the debate about the neutrality of metaethics and the attendant claim about the autonomy of morality. This objection applies because one might take theories of moral status to be irreducibly normative ethical theories and using them in the way proposed seems to entail that metaethics is decidedly not neutral with regard to normative ethical matters. As I note above, theories of what constitutes moral status—say, whether it is sentience or personhood—are irreducibly normative insofar as they claim that a specific set of facts both have a moral valence and establish moral obligations. But claims about whether there are facts that underwrite moral status, what their metaphysical status is, and how we know them are normative ethically neutral.
With these objections addressed, I now consider how to use the Ecumenical View of Moral Status to vindicate naturalist moral realism. To do so, I need to translate the Ecumenical View of Moral Status (and the two attendant Propositions discussed in the previous section) into an independent premise, which I call the Moral Status Premise:
1. Moral Status Premise: If others (creatures, entities, and the like) are to have moral status, there must be some characteristics that they actually possess or could possess that ground, fix, or constitute such status.
2. For such facts to do so, they must have a relationship with a set of evaluative facts, either as the latter's subvenient or reduction base.
3. If such facts exist, then some form of (reductionist or nonreductionist) naturalist moral realism is true.
4. Therefore, if others (creatures, entities, and the like) are to have moral status, then some form of (reductionist or nonreductionist) naturalist moral realism is true.
This argument shows that the Moral Status Premise can be used to vindicate the naturalist moral realist position. For the sake of simplicity, I leave aside nonnaturalist moral realism; however, one could easily remove reference to the ‘natural facts’ that underwrite moral status in the argument above such that it could be used to vindicate nonnaturalist forms of moral realism as well.
But does the argument above show that the naturalist moral realist can justifiably use the Moral Status Premise to vindicate her view? It would seem that she can, by extending the argument above as follows:
5. There are some creatures who have moral status.
6. Therefore, some form of naturalist moral realism (either a reductionist or nonreductionist variety) is true.
Unfortunately, the Moral Status Premise makes arguments either for moral realism or against noncognitivist, antirealism question-begging. First, it requires, in combination with premise (5), that we must assert that there are indeed some creatures who have moral status. A moral realist view cannot assert that there are indeed moral facts, such as those that bestow moral status, in order to prove that moral realism is true. Rather, a moral realist can only justifiably use the Moral Status Premise in such a way that grants that there may be beings who possess moral status. Consider an analogy: I can say what it would be for a being to have free will even if (a) no actual beings in this world or nearby possible worlds have free will or (b) no circumstances in this world and nearby possible worlds make it (or could make it) such that free will is possible.
Second, there is a related problem in premises (2)–(3). The combination of these two premises and the Moral Status Premise is also question-begging against noncognitivist antirealism. This is because the combination of these claims assumes—without argument—that we cannot both grant that there are facts that ground moral status and deny that if there are facts that ground moral status, they must serve as the mind-independent reductionist or subvenient base of the evaluative facts constituting moral status.
In terms of the first question-begging feature of the argument, we ought not drop the Moral Status Premise altogether; rather, we ought to modify how it is used. The Ecumenical View of Moral Status should not be a premise in proving a metaethical view true or false. Instead, it should be an outer constraint on what sorts of views are plausible views to consider. So it serves a constraining function, and I rename it the Moral Status Constraint. In terms of the second question-begging feature of the argument, we need to modify the kind of constraints it places on metaethical views—ones that are neutral with regard to realist or antirealist explanations of moral status. Doing so is consistent with my claims above regarding the sense in which a view of moral status should be ecumenical. So we should reformulate the Moral Status Constraint, as it relates to the plausibility of metaethical views more generally, as follows:
Any view about the existence of moral facts—whether to deny or to defend it—must be consistent with a subset of the following claims:
a. There are or could be creatures to whom we owe moral consideration and thus they possess moral status.
b. This status is or could be fixed by a set of relevant characteristics. These characteristics need not be understood as themselves moral facts, although they must be descriptive facts of the matter. [The Proposition that Moral Status is Mind-Independent and The Proposition that Moral Status is Grounded in Natural Facts]
c. In cases where the characteristics are not fixed by facts, there is or could be a clear procedure for identifying these characteristics—that is, if one gains this status, say, by virtue of being a contractor or the addressee of a moral claim. [The Proposition that Moral Status is Mind-Independent and The Proposition that Moral Status is Grounded in Natural Facts]
d. Any metaethical explanation of moral status must remain consistent with and explain the possibility of actual genuine (first-order normative) disagreement about the facts that constitute moral status. [To remain ecumenical regarding normative ethical theories]
To be consistent with the Moral Status Constraint, a metaethical view must be consistent with (a) and (d). It must then be consistent with [(b) or (c)]. More demandingly, a view that is able to provide an independent explanation of (a), (d) and [(b) or (c)] will provide a better account of the Moral Status Constraint than those that are merely consistent with it.
I consider briefly the motivation for each of these conditions, with a particular focus on how they differ from the question-begging Moral Status Premise and the attendant argument in which it was used. Condition (a) is straightforward, given that metaethical views must grant that there are, in principle, creatures that have moral standing in order for the possibility of moral status to be a useful test of the view's plausibility. Moreover, as noted earlier, most views, with the exception of some error theories, will grant (a). What they will disagree about is how such status is constituted or granted.
In order to remain fully metaethically ecumenical, as well as to avoid the question-begging combination of premises (2)–(3) in the argument above, I have added the disjunctive conditions (b)–(c). Each of these conditions is merely a restatement of the conjunction of the Propositions that Moral Status is Mind-Independent and that Moral Status is Grounded in Natural Facts, which, as noted in the previous section, are consistent with a wide variety of metaethical views including noncognitivist antirealist views.
It is (d) that is a surprising addition. To highlight the motivation for adding it, notice that there are two ways in which the Moral Status Constraint should remain ecumenical regarding moral status. First, it should not implicitly favor one metaethical view by assuming the truth of claims about moral facts or moral knowledge (among others) that are sufficiently similar to those of one particular metaethical view. Second, it should be consistent with the neutrality of metaethics and the autonomy of normative ethics,Footnote 9 where this autonomy requires (minimally) that metaethical views do not block or purport to solve the actual, genuine first-order normative disagreement about what constitutes moral status. Notice that this need not commit us to any claim about the general autonomy of morality thesis; rather, if the Moral Status Constraint is a test for the plausibility of a metaethical view, the view in question should be consistent with the fact that the question of moral status is unresolved and that it is a matter of significant first-order debate.Footnote 10
The constraining function discussed above seems to be a promising route for using the Ecumenical View of Moral Status to vindicate naturalist moral realism in a non–question-begging fashion. If we have independent reasons for thinking that there must be some characteristics that constitute or ground moral status, then we have a reason to use it as a constraint in evaluating the truth of metaethical views.
2. Why the Moral Status Constraint Vindicates Naturalist Moral Realism
My aim in this section is to establish the extent to which naturalist moral realism is able to cope with this constraint in comparison to noncognitivist antirealist views. The model for this approach is the same one that motivated this paper—namely, to find an analogue to vindicate naturalist moral realism in light of how the Humean theory of motivation and judgment internalism are used to defend noncognitivist antirealism. With this point in mind, my purpose here is not to show that we should be absolutely certain about the untenability of other views, although note that some metaethical views may be ruled out simply by virtue of the nature of the view of moral status (however ecumenical it is). Most error-theoretic views as well as broadly nihilist views may be unable to be made consistent with such a view, unless, for example, they treat moral status as a socially useful, but not a truth-tracking, moral notion.Footnote 11 My purpose in what follows is to show that views that fit better with the Moral Status Constraint have an advantage, given that they explain features of our moral life that other views cannot.
2.1. Moral Realism's Explanation of The Moral Status Constraint
Both naturalist and nonnaturalist moral realist views would seem to cope well with the Moral Status Constraint. But I would like to focus on naturalist moralist realism in particular. Here I choose a reductionist version of moral realism, as it is the clearest example of a moral realist view that can do so. This choice is dictated largely by space constraints, and not because I think that nonreductionist (or ‘Cornell’) realist views have a serious problem explaining the Constraint. In fact, one might think that they have an easier route in that they can plausibly explain (d), or the disagreement condition, by appealing to questions concerning the nature of the supervenience relationship between moral status facts and their subvenient (or ‘resultance’) base. In this regard, differences among forms of moral realism should not affect their ability to meet the Moral Status Constraint, although they may affect the way by which they do so. One could just as easily choose an exemplar of Cornell realism, such as that defended by Brink (Reference Brink1989) or Sturgeon (Reference Sturgeon and Sayre-McCord1988; Reference Sturgeon and Dreier2006), or focus on the naturalism defended by Copp (Reference Copp2007: especially essay 5), for the sake of determining how the moral realist will cope with the Moral Status Constraint. Naturally, moral nonnaturalists, such as Shafer-Landau (Reference Shafer-Landau2003), would defend the Moral Status Constraint in superficially different terms than either the Cornell realists or reductionist naturalist moral realists. The main difference here would be the metaphysical status of the facts at hand, not whether there are such facts to be had.
The example of reductionist naturalist moral realism I have in mind is Peter Railton's (Reference Railton2003a; Reference Railton2003b) version of full information accounts of the good or ideal advisor views. Full information accounts of the good have been understood in a variety of ways, as exemplified by those versions defended by Firth (Reference Firth1952), Sidgwick (Reference Sidgwick1907 [1981]), Brandt (Reference Brandt1955; Reference Brandt1979) and Smith (Reference Smith1994; Reference Smith1995). For simplicity's sake, I choose Railton's version because it is paired with a version of naturalist moral realism, which is the type of view currently under consideration. On Railton's view, moral facts are simply a special set of natural facts about the kinds of ways that we can flourish, given the kinds of natural and social creatures that we are. We can come to know them by appealing to the perspective of an idealized advisor, who has access to full information about our circumstances and represents the perspective of what is morally good for individuals as a group. Doing so allows us to know what is morally good from what Railton (Reference Railton2003a: 21) calls ‘the social point of view’.
Although much can be said about Railton's view (e.g., Arkonovich [Reference Arkonovich2011]; Kawall [Reference Kawall2006]; Rosati [Reference Rosati1995]; Sobel [Reference Sobel1994];Tiberius [Reference Tiberius1997]), my aim here is to show how an exemplar of moral realism might account for the Moral Status Constraint. Consider how a proponent of a Railton-esque moral realist view would address the aforementioned conditions (a)–(d). It is clear that she can easily meet constraint (a), given that she grants that moral facts exist. What is more interesting is how the moral realist will meet (b). The Railton-esque moral realist must provide an explanation of how a set of descriptive facts of the matter—say, one's capacity to suffer, to be harmed by cruelty, among other facts about human beings—are the proper reduction basis for moral status (Railton Reference Railton2003a: 10, Reference Railton2003b: 62–63). A Railton-esque moral realist can do this by identifying, from the perspective of the generalized ideal advisor, what sorts of beings are those for which there is a moral good. In doing so, she explains, as Railton (Reference Railton2005: 430) puts it, ‘moral properties … in non-moral terms via largely functional “job descriptions” of their putative role in thought, self-regulation, and social practice’.Footnote 12 She thus identifies the natural, common characteristics that underwrite—or serve as the ‘reduction basis’ (Railton Reference Railton2003b) for—their moral status. In this regard, the device of the ideal advisor provides both the metaphysical grounding for the claim that these are indeed moral facts that define moral status and epistemic grounding for the claim that we can have defeasible moral knowledge about them. It is worth noting again that different types of moral realists will explain the nature of these facts in different ways—here consider how a nonreductionist naturalist might do so. Still, the way by which they do so has the same structure. Namely, they cite the facts that underwrite moral status, where moral status is reduced to those facts or supervenes on them.
Interestingly enough, the Railton-esque view will have the most difficulty meeting (d), although I contend that this view (and related naturalist moral realist views) nonetheless succeed(s). One would think that it should be easy for a moral realist to admit that we can disagree about which moral facts constitute moral status, given that they admit that we may be mistaken about what the moral facts are. But notice that this admission does not fully meet the target explanandum in (d), given how the moral realist has explained the other conditions. Rather, to explain disagreement about first-order theories of moral status, a naturalist moral realist must explain the following:
1. The actual disagreement about the facts (say, the capacity for suffering versus personhood) that constitute moral status.
2. How it is possible for us to have different, competing ways of moving from (or translating) natural facts into moral facts such that we disagree about which route is appropriate.
As I have noted, explaining (1) is relatively easy for the naturalist moral realist. It is (2) that presents a more serious problem. This is because the moral realist must explain how the translation mechanism (or, for the nonreductionist, the proposed supervenience relationship) can be plausibly understood in different, competing ways even while granting the existence of moral facts that arise from them. So, in the case of disagreement that I am imagining, two people plausibly agree that there is such a thing as moral status. They disagree about the natural facts that underwrite it and about how one moves from these natural facts to moral facts about a creature's moral status.
Here the naturalist moral realist can appeal to disagreement about how the proper reduction from moral facts to natural facts should occur, and what the metric for such a reduction should be. In this regard, the naturalist moral realist can show that two people can agree that there is a moral question about moral status to be answered while plausibly disagreeing about how precisely the natural facts line up to underwrite such status.
3. Quasi-Realism, Moral Status and a New Problem Explaining Disagreement
If moral realists claim that their consistency with and ability to account for the Moral Status Constraint is an additional reason to think that moral realism is true, then we must show that noncognitivist antirealist views fare less well on this front. Since the model for my argument in this paper is the noncognitivist antirealist's use of the Humean theory of motivation to vindicate her view, the comparison should be between moral realism's ability to account for or meet the Constraint and expressivism's ability to do so.Footnote 13 I leave aside what are often called ‘cognitivist’ forms of expressivism (e.g., Barker Reference Barker2010; Copp Reference Copp2001; Horgan and Timmons Reference Horgan, Timmons, Horgan and Timmons2006). As Copp (Reference Copp2001: 155–56) suggests in his discussion of what he calls ‘realist-expressivism’, these views can accommodate any form of moral realism and are thus more properly realist views than expressivist ones. In order to test how well expressivists views cope with the Moral Status Constraint in comparison to moral realist views, I choose to focus on those versions of the view that are (relatively speaking) farther away from moral realism—namely, emotivism and quasi-realism.Footnote 14
Consider whether two forms of expressivism are able to meet the Moral Status Constraint. Broadly speaking, emotivists will have the hardest time meeting it. They can acknowledge that there are or could be beings who possess moral status, thereby meeting (a) of the Constraint. But they cannot grant or explain (b)—that this status is or could be fixed by a set of relevant characteristics that are mind-independent. Since they cannot imagine (b), then they must provide an answer to (c). Emotivists are unlikely to be able to meet (c), given that they cannot provide a clear procedure for identifying those beings who possess moral status. This is, in part, due to the fact that there is no reliable set of criteria for how to use emotive utterances to represent moral sentiments (Ayer Reference Ayer1952; Schroeder Reference Schroeder2008). The reason for this is simple: if moral claims are simply emotive utterances and there is no strong criterion for what constitutes an appropriate emotive utterance, there will be no way to generate a generalizable procedure for identifying the characteristics that underwrite others’ moral status. Here one might note that we may be able to identify a general pattern in our psychological tendencies to attribute moral status to others (and ourselves).Footnote 15 Such an account could go some way in solving this problem for emotivism, although it will not enable it to compete with the quasi-realist's more robust explanation of (c), as I describe below.
By contrast, a more promising form of expressivism, at least with regard to the question of the Moral Status Constraint, is quasi-realism. This is because quasi-realists argue that we can earn the right to talk as though there are moral facts of the matter even while we deny that they are mind-independent (Blackburn Reference Blackburn1984; Reference Blackburn1998; Reference Blackburn1993b; Reference Blackburn1993a; Gibbard Reference Gibbard1990; Reference Gibbard2003; but compare with Dreier Reference Dreier2012; Harcourt Reference Harcourt2005; Ridge Reference Ridge2006; Schroeter and Schroeter Reference Schroeter and Schroeter2005). The most prominent examples of such views are Blackburn's quasi-realism and Gibbard's norm-expressivism. I briefly consider each view in turn. For Blackburn (Reference Blackburn1984: 197, 216–18; Reference Blackburn1988; Reference Blackburn1993b: 124–25; Reference Blackburn1993a: 168; Reference Blackburn1998: 4–5, 51–53; Reference Blackburn2015: 846–47), moral claims are the product of a naturalistic, causal process. Blackburn (Reference Blackburn1998: 59) puts the point as follows:
Expressivism requires a naturalistic story of the state of mind of valuing something. We then go on from that to give an account of the procedures of valuation that we adopt; the modes of expression that are appropriate; and finally the logic and theory of meaning of our typical expression of values.
Blackburn's version of quasi-realism takes on the challenge of earning the right to talk as though there were moral truths by proposing a naturalistic account of how we come to value particular things. On this view, the states of mind that represent how we come to value particular things are ethical sensibilities. These sensibilities provide the ground for the moral claims that we express. Still, these moral claims are appropriate or inappropriate responses to states of affairs. They are appropriate if the states of mind that give rise to them represent outputs that are properly responsive to the inputs provided by states of affairs (Blackburn Reference Blackburn1998: 4–5).
Since the outputs are ethical sensibilities, they are not truth-tacking responses to a set of moral facts, contra the moral realist. These sensibilities are mind-dependent modes of valuing or disapproving of states of affairs. Nonetheless, these outputs are the product of a naturalistic process that approximates, to more or less accurate degrees, the features of the scenario towards which it is appropriate to have a moral response. As a result, we can, Blackburn thinks, talk as realists without granting that the moral qualities that we assign to states of affairs actually inhere in them. In this regard, Blackburn (Reference Blackburn2015) emphasizes the affinity between the upshots of a Cornell realist's view and his version of expressivism. Still, for the sake of comparison, it is worth noting that one can defend a similar kind of claim that earns for quasi-realists the right to talk as realists but without the background naturalistic story (Street Reference Street and Shafer-Landau2011). Allan Gibbard's (Reference Gibbard2003) defense of his norm-expressivism provides just this alternative. Gibbard (Reference Gibbard2003: 168–69) suggests that we earn the right to talk as realists when the moral terms that we use play a central role in sound practical reasoning: ‘For a term to signify the brutal as a genuine normative kind term, it must play the right role, more or less, in its users’ normative thinking and discourse. It must figure, more or less correctly, in explanations of what to do or how to feel about things’.
Notwithstanding the fact that Blackburn's quasi-realism and Gibbard's norm-expressivism differ with regard to the question of whether moral expressions are naturalistically or nonnaturalistically grounded, both views will have a similar approach to the Moral Status Constraint. Here, the quasi-realist's route to securing the right to ‘talk as realists’ will play a central role. Namely, it must be sufficient for fixing the characteristics that underwrite moral status—condition (a)—and it must provide a clear route to meeting either condition (b) or (c). As will become apparent, the quasi-realist has some trouble meeting both conditions. But even granting that she can meet them, the real problem for the quasi-realist arises with (d), where she has familiar trouble explaining moral disagreement but for new, unfamiliar reasons.
On a generous reading, the quasi-realist can meet (a) by acknowledging that there are indeed facts about others that secure their moral status. But what makes their status specifically moral, the quasi-realist will say, is that we respond to them with the relevant moral attitudes or practically normative judgments. To see why, consider in greater detail how she understands the route to securing the right to talk as realists. Although there are a number of reasons to worry that this approach cannot secure this right (e.g., Schroeder Reference Schroeder2008; Street Reference Street and Shafer-Landau2011), I treat the view as though it were successful in doing so, given that the real question is whether securing this right provides the route to explaining the Moral Status Constraint. Street (Reference Street and Shafer-Landau2011: 9) describes what securing this right actually involves:
One might seem to be suffering from a split personality if one claims, when wearing one's hat as a metaethical theorist, that to say ‘Kicking dogs is wrong’ is to express a state of mind of planning, and yet insists, when wearing one's hat as a participant in normative discourse, that ‘Kicking dogs is wrong completely independently of us and our planning’. Quasi-realists forcefully argue, however, that on a proper understanding of their view, there is no tension whatsoever involved in simultaneously holding the two positions. The latter position, they explain, is a normative judgment, whereas the former is a meta-level claim explaining what we're doing when we make a normative judgment, and these two levels of claim and understanding do not interact.
What makes quasi-realism unique is that while its proponents seem to deny that moral truths are mind-independent, they actually make a more nuanced claim. They argue, first, that when we make moral claims, we are expressing attitudes of approval or disapproval (for Blackburn) or normative planning (for Gibbard). But that this is what we do when we make moral claims is not inconsistent with the fact that we also make what Street calls ‘normative judgments’ about the status of those moral facts. We suppose—mistakenly—that both types of expressions are mind-dependent. In fact, however, only the first type of expression is mind-dependent, while the second is a mind-independent claim about the status or possibility of moral truth. So quasi-realism can acknowledge that there are (or could) be creatures who possess moral status, where ‘having’ moral status is constituted by the role the relevant attitude plays in moral reasoning.
Recalling Street's description above, however, there appears to be a problem. In making a moral claim, we do two things: we hold attitudes by which we earn the right to talk as though there were mind-independent facts that constitute moral status and we explain the status of these moral claims. If this is right, it is unclear whether quasi-realists succeed in explaining in what moral status consists, which meeting (a) requires. Yet, as Street notes, the quasi-realist will argue that there is no tension here: the claim about moral status is a moral claim while the other is a ‘meta-level’ claim about how we are justified (or not) in asserting it.
With this said, there are still reasons to think that the quasi-realist may struggle with (a) if pushed on grounds similar to the ones just outlined. Consider Dreier's (Reference Dreier2004: 38-9) claim that, for the quasi-realist, moral judgments serve two functional roles: both fixing moral facts and, simultaneously, constituting the moral judgments about those facts. If this is right, then it is a potential problem for quasi-realism's ability to explain (a) of the Moral Status Constraint. The quasi-realist wants the appropriate attitudes in response to neutral facts of the matter to play two, simultaneous roles that it cannot plausibly play: namely, (1) to provide the relevant moral patina for these neutral facts; and (2) to constitute the very moral judgments that we would think typically follow from (or are made in light of) these facts.
With this said, let us give the quasi-realist the benefit of the doubt regarding (a). Given that meeting the Moral Status Constraint requires more than meeting (a), however, I now consider whether quasi-realists will choose to meet either (b) or (c). The quasi-realist can meet (c) by providing an account of how we make normative judgments about others as deserving of moral consideration, as my discussion of Street's (Reference Street and Shafer-Landau2011) gloss on quasi-realism makes clear.Footnote 16 This procedure is either largely naturalistic (for Blackburn) or determined by the place of normative judgments in good practical reasoning (for Gibbard). In both cases, we identify a distinct procedure that results in moral status judgments, thereby meeting (c).
Here one might worry that while the quasi-realist can identify a distinct procedure for the formation of evaluative attitudes, it may not be a causally reliable one. Mark Schroeder (Reference Schroeder2008: 26–27) has raised concerns along these lines, noting that the quasi-realist must appeal to a causal theory about the relationship between the state of affairs and our beliefs (or our belief-evaluative expressive attitudes).Footnote 17 Yet, in doing so, she will also have to explain the place of insincere expressions in this causal process, which is a well-known difficulty for quasi-realists and expressivists more generally. It will thereby make it difficult for the quasi-realist to explain how the same causal process produces both sincere and insincere attitudinal responses. And, as result, the causal process that is essential for the argument appears to be unreliable in that it does not guarantee, under normal circumstances, the formation of the relevant moral attitudes. From the perspective of explaining (c) of the Moral Status Constraint, one might worry that the quasi-realist cannot explain the reliable process by which we secure the right to make, in this case, moral status claims.
Here I think that we can defend the quasi-realist, at least with regard to the question of whether she meets (c). Notice that the bar for meeting (c) is quite low—a metaethical view must simply identify a procedure, where the procedure in question makes a contentful contribution to the constitution of moral status. Nothing about the way that this condition is formulated requires that this procedure be reliable. In this regard, the type of procedure that the quasi-realist needs to explain is not the causal process by which agents form the relevant moral judgments. Rather, it is the procedure by which the relevant judgments render morally neutral facts into morally relevant ones. For the quasi-realist to do this, she merely needs to explain why moral attitudes are morally expressive.
As I note above, although we may raise doubts about quasi-realism's ability the meet the other conditions of the Moral Status Constraint, the real problem arises in its ability to meet (d). What is noteworthy here is that although quasi-realism struggles to meet (d), it is not for the familiar reasons. The problem of explaining disagreement is a well-known issue for expressivism, and it affects different versions of expressivism in different ways.Footnote 18 Broadly speaking, expressivists have a problem explaining disagreement because there is no sense in which if I say ‘Stealing is wrong’ and you say ‘Stealing is not wrong’, we are genuinely asserting what we each understand as competing, mutually inconsistent propositions that have the same object of description.Footnote 19 There is significant debate about the nature of this problem, and I do not intend to solve it here. While the aforementioned version of the problem of disagreement may be a genuine one,Footnote 20 it is not the problem that affects the quasi-realist's ability to account for (d), notwithstanding the fact that both involve disagreement in some sense of the term. If Sophia utters the phrase ‘Suffering is necessary and sufficient for moral status’, and Juan utters the phrase ‘Personhood is necessary and sufficient for moral status’, the kind of disagreement we must explain is different in kind from the stealing case. This is not to deny that these are in principle competing propositions. Rather, the type of disagreement and its proper domain are distinct in each case. In the first type of disagreement, what has to be explained is how agents can both treat their utterances regarding stealing as expressive and yet grant that the propositions are competing. In the second case, what has to be explained is how agents can understand that they have each drawn different connections between morally neutral facts and their expressive claims regarding how those facts constitute or establish moral status.
Consider the argument in more detail. Recall that, to explain the relevant sense of disagreement in (d), the naturalist moral realist has two explananda. The quasi-realist also has two analogous explananda:
1. The fact that two agents may have (potentially) competing first-order normative ethical judgments about which facts ground moral status.
2. How it is possible for two agents to have different (in principle competing) ways of moving from (or translating) inputs for moral expressions to the expressive utterances themselves, such that they differ not only regarding the propriety of the latter but about how one should move from the former to the latter.
It is unclear whether quasi-realists can meet (1), given their aforementioned problem explaining disagreement. But notice that (1) does not require explaining how Sofia and Juan disagree in the standard sense of the term; rather, it requires explaining how they can make the relevant connections between morally neutral facts about others and their expressive, first-order claims regarding how those facts constitute or establish moral status. As long as the quasi-realist can plot out the relationship between morally neutral facts and expressive attitudes, she has met this burden.
It is (2) that is the problem for the quasi-realist. Quasi-realists (and expressivists more generally) cannot explain the route by which individuals arrive at accounts of moral status that they understand as such. This is because it is unclear from which perspective Sofia would be able to reconstruct how Juan came to form his expressive attitude regarding moral status. Juan would encounter the same problem as Sofia. Why? Per quasi-realism, expressive attitudes concerning moral status are responsive to morally neutral facts of the matter (whether naturalistically or with reference to a picture of sound normative reasoning). If this is so, then it is impossible for them to each understand each other's respective routes to forming their views of moral status as such. Thus, notwithstanding the fact that the disagreement in question is not the traditional sort that causes problems for expressivists, the quasi-realist cannot meet the Moral Status Constraint because she cannot fully explain first-order disagreement about moral status.
4. Conclusion
It would seem, then, that the naturalist moral realist can meet the Moral Status Constraint. By contrast, her closest competitor—the quasi-realist—struggles to meet (a) and (c) of the Constraint. But even if we were to grant that the quasi-realist can meet these two conditions, the real problem for the quasi-realist lies in her inability to meet (d). For these reasons, it appears that we have a plausible route to vindicating naturalist moral realism, notwithstanding genuine first-order disagreement about the very concepts that we are using to do so.
There are certainly other metaethical views that would likely also cope well with this constraint, but it is impossible to discuss all of them here. All that I have shown is that naturalist moral realism provides a more plausible account of the Moral Status Constraint than one of its stronger contenders. In this way, like the use of the Humean theory of motivation and theories of moral motivation to vindicate noncognitivist antirealism, we have a structurally comparable route to vindicating naturalist moral realism. More weakly, we have what Enoch (Reference Enoch2011: 15) calls a ‘plausibility point’ in favor of naturalist moral realism. But even if the reader is unwilling to grant these points, she should recognize that we have at least added another potentially vindicatory concept to our metaethical arsenal. Just as moral motivation is an indispensable moral concept with which metaethical accounts must grapple, so, too, should they grapple with a relatively ecumenical theory of moral status.
