There’s Nothing Quasi About Quasi-Realism: Moral Realism as a Moral Doctrine

Abstract

This paper seeks to clarify and defend the proposition that moral realism is best elaborated as a moral doctrine. I begin by upholding Ronald Dworkin’s anti-Archimedean critique of the error theory against some strictures by Michael Smith, and I then briefly suggest how a proponent of moral realism as a moral doctrine would respond to Smith’s defense of the Archimedeanism of expressivism. Thereafter, this paper moves to its chief endeavor. By differentiating clearly between expressivism and quasi-realism (or moral realism as a moral doctrine), the paper highlights both their distinctness and their compatibility. In so doing, it underscores the affinities between Blackburnian quasi-realism and moral realism as a moral doctrine. Finally, this paper contends—in line with my earlier work on these matters—that moral realism as a moral doctrine points to the need for some reorienting of meta-ethical enquiries rather than for the abandoning of them.

This paper takes as its point of departure an article published by Michael Smith in the 2010 volume of the Boston University Law Review. That article responds to Ronald Dworkin’s critique of the error theory of morality and his critique of expressivism. Although this paper will seek to explain why Smith’s retorts to those critiques are unsuccessful, it will also criticize Dworkin in some respects. While I largely agree with Dworkin’s conclusions about the error theory, I will contend that his argument against it moves too quickly. Some further argumentation is needed, which this paper will endeavor to provide. In the second half of the paper, I will move even further away from Dworkin. Although he was right to think that expressivism as he understood it is a moral doctrine, there is an alternative version of expressivism that is not in itself such a doctrine. In maintaining as much, I will neither be siding with Smith nor be retreating in any other way from the view—which I share with Dworkin—that meta-ethical matters are ethical matters (usually at very high levels of abstraction). There will be no such retreat, because expressivism properly understood is not about morality but is instead about the activity of moral discourse. It is a philosophical account of the pragmatics of such discourse. Accordingly, it is entirely consistent with moral realism as a moral doctrine.

Because the first two principal sections of this paper (especially the first principal section) may appear somewhat disconnected from the remaining portions, I should here briefly clarify how the paper hangs together. Moral realism as a moral doctrine advances two main contentions: first, the claim that meta-ethical matters are ethical matters; and second, the claim that morality is strongly objective in a number of different respects. Let us designate the first of those contentions as the “Reconception of Meta-Ethics Proposition” and the second of them as the “Objectivity Proposition.” In the opening section of this paper, the focus lies on the Reconception of Meta-Ethics Proposition—as the paper seeks to bolster Dworkin’s effort to show that the error theory developed by John Mackie is an ethical doctrine that carries certain ethical implications. Although the focus in Sect. 2 of the paper also lies on the Reconception of Meta-Ethics Proposition, that section is a transition to the paper’s subsequent discussions. Instead of amplifying and defending Dworkin’s endeavor to show that expressivism as a meta-ethical account of the semantics of moral assertions is an ethical theory with ethical implications, I will tersely recount his reasoning and Smith’s response thereto. Though my short summary will be sympathetic to Dworkin’s position, its chief purpose is to set the stage for an exploration of a markedly different version of expressivism that is not similarly within the sway of the Reconception of Meta-Ethics Proposition. That latter version of expressivism is recounted in Sect. 3 of the paper, which will maintain that expressivism construed as a philosophical account of the pragmatics of moral discourse is fully consistent with both the Reconception of Meta-Ethics Proposition and the Objectivity Proposition. Expressivism so construed is therefore fully consistent with moral realism as a moral doctrine. After developing that central point, Sect. 3 will further elucidate moral realism as a moral doctrine by highlighting the centrality of a minimalist conception of truth within it and by emphasizing that it calls for a reconception—rather than a repudiation—of meta-ethics. The short concluding section of the paper, Sect. 4, laconically recapitulates these several strands of my discussion.

Dworkin and Smith on the Error Theory

In the first half of his article, Smith concentrates on a passage by Dworkin that deals with Mackie’s version of the error theory. That passage was contained in an unpublished draft of Dworkin (2011), but—partly in response to Smith (2010)—Dworkin somewhat modified his discussion of the error theory in the published version of his book. (Dworkin 2011: 32–37, 40–51) Still, because this section of my paper will be focusing principally on Smith’s rejoinder to Dworkin, and because in any event Dworkin did not alter the gist of his critique of the error theory, I will quote here the passage by Dworkin that appears in Smith’s article:

There are three available positions about most questions of moral right and wrong. We may believe that affirmative action programs are morally wrong, or that they are morally required, or that they are neither wrong nor required. If they are neither required nor morally wrong, which is what many educators think, then a university is morally free to make a judgment one way or the other on non-moral grounds: to serve optional educational goals, for example. Presumably that is also the claim that an external error skeptic makes about affirmative action, since he rejects the alternative claims both that affirmative action is morally forbidden and that it is morally required. But the proposition that affirmative action is neither required nor forbidden is plainly itself a moral judgment. Educators who take that view offer moral arguments in its favor by challenging, on substantive moral grounds, the claims of both those who insist that it is forbidden and those who say it is required. The external skeptic must claim that some moral judgments, including that permissive one, are true. (Dworkin 2009: 42, quoted in Smith 2010: 511–512)

Smith’s Response to Dworkin’s Critique of the Error Theory

Smith responds to this quoted passage by asserting that Dworkin failed to recognize that error theorists not only deny that any modes of conduct are ever morally forbidden or morally obligatory; they also deny that any modes of conduct are ever morally permissible. (Smith 2010: 512) However, when Smith endeavors to substantiate this claim about the error theory, his argument is quite unsuitable. On the one hand, he correctly points out that Mackie’s version of the error theory construes goodness and wrongness and various other ethical properties as objectively prescriptive. Mackie deemed the combination of objectivity and prescriptiveness to be incoherent. Having noted as much, Smith has explained why the error theory denies that an affirmative-action policy is ever morally required or morally forbidden. On the other hand, however, he has not gone any way toward explaining why the error theory would deny that such a policy is ever morally permissible—for, unlike the properties of moral obligatoriness and moral forbiddenness, the property of moral permissibility is not prescriptive in Mackie’s sense.

Mackie took objective prescriptiveness to consist in to-be-pursuedness or not-to-be-doneness. (Mackie 1977: 40) That is, a morally obligatory course of conduct is endowed with to-be-pursuedness because of the objective nature of that course of conduct which makes it obligatory, rather than because of anyone’s aims or desires. Someone who recognizes the moral obligatoriness of that course of conduct—its to-be-pursuedness—will thereby feel inclined to act in conformity to it. Conversely, because of the objective nature of a morally forbidden course of conduct, rather than because of anyone’s aims or desires, such a course of conduct is not-to-be-done. Someone who recognizes the moral forbiddenness of that conduct will thereby feel disinclined to engage in it. So Mackie contended, as he also maintained that objective to-be-pursuedness and objective not-to-be-doneness are too queer to be countenanced.

Noteworthy here is that the property of moral permissibility does not imbue a course of conduct with either to-be-pursuedness or not-to-be-doneness. An ascription of moral permissibility to some mode of conduct (MC) directly denies that MC is possessed of not-to-be-doneness, and it leaves entirely open the question whether MC is possessed of to-be-pursuedness. Hence, such an ascription does not attribute prescriptiveness to MC at all. Consequently, the queerness of objective to-be-pursuedness and objective not-to-be-doneness—or the incoherence of the notion of objective prescriptivity—is no basis whatsoever for error theorists to deny that countless modes of conduct are morally permissible (which may be why Mackie himself focused on goodness and wrongness and not on permissibility). Smith has gone astray in suggesting otherwise.

This riposte to Smith is equally pertinent in application to the elegant re-articulation of the error theory by Jonas Olson in his recent book Moral Error Theory. For Olson, moral facts are queer because “they are or entail facts that count in favour of or require certain courses of behaviour, where the favouring relation is irreducibly normative.”¹ The fact of MC’s moral permissibility obviously does not mean that MC is required, nor does it even count in favor of MC; it simply does not count against MC. An ascription of moral permissibility to MC does not in itself indicate that there is any reason which favors the adoption of MC, and it leaves open the possibility that the only reasons pertaining to MC in any particular context are such as to count conclusively against the adoption of that mode of conduct. The sole purport of the ascription is to indicate that a powerful reason against the adoption of MC—the moral obligatoriness of eschewing such a mode of conduct—does not exist. Since error theorists themselves insist that no reason of moral obligatoriness ever exists, an ascription of moral permissibility is fully consistent with their doctrine. (Even an extreme act-utilitarian creed, which affirms that no course of conduct is morally permissible unless it is maximally promotive of utility, has to allow for multiple optima. Hence, even under the terms of such a creed, an ascription of permissibility does not entail an ascription of obligatoriness. In any event, an error theorist is scarcely in a position to assume the correctness of any extreme act-utilitarian tenets, nor could an error theorist plausibly maintain that the correctness of those tenets is presupposed by all ordinary moral judgments.)

An Alternative Argumentative Strategy?

The place of permissibility in Mackie’s error theory is linked to a highly underdeveloped feature of that theory. On the one hand, Mackie at times implied that all moral judgments by ordinary people are false. He wrote, for example, that “although most people in making moral judgements implicitly claim, among other things, to be pointing to something objectively prescriptive, these claims are all false.” (Mackie 1977: 35) On the other hand, in his more considered moments Mackie adopted a somewhat milder position. He contended not that all moral judgments are false, but instead that “all affirmative first order [moral] judgements are false.” (Mackie 1977: 49) I shall here follow the present-day error theorist Charles Pigden in substituting “logically atomic” for “first order” and in thus taking “affirmative” to be a pleonastic means of emphasizing that negated ascriptions of moral properties to modes of conduct can be true. (Pigden 2007: 451) Indeed, the very coherence of the error theory depends on the truth of countless such negated ascriptions. Error theorists cannot coherently maintain that the proposition “Torturing babies for pleasure is morally wrong” and the proposition “It is not the case that torturing babies for pleasure is morally wrong” are both false. Error theorists have to deem the latter proposition to be true.

However, this feature of the error theory gives rise to some major problems. In the present context, the most notable problem is that “It is not the case that torturing babies for pleasure is morally wrong” appears to entail “Torturing babies for pleasure is morally permissible.” Hence, error theorists appear to be committed to the truth of this permissibility-ascribing proposition as well as of countless other permissibility-ascribing propositions. Mackie’s unelucidated restriction of the scope of the error theory to affirmative moral judgments does appear to carry the implications which Dworkin imputed to it.

Is there any escape from this problem for the error theorists? An appeal to the queerness or incoherence of objective prescriptiveness will be of no avail, for, as has been observed, permissibility-ascribing propositions do not presuppose that there are any objectively prescriptive properties or principles. Error theorists need to adopt some other tack. I will broach here a route that might be pursued by them, though I will conclude that any such pursuit would be unsuccessful.

Dworkin in effect submitted that every x is morally forbidden or morally obligatory or morally permissible-but-not-obligatory. Over what domain does the universally quantified “x” variable range? Quite plainly, that domain does not comprise all events or processes or states of affairs. For example, if we inquire whether the orbiting of the planet Jupiter around the sun is morally forbidden or morally obligatory or morally permissible-but-not-obligatory, we should conclude that it does not partake of any of those moral properties. That orbiting is obviously not morally obligatory or forbidden, but it is likewise not morally permissible. No moral properties are appositely predicable, for the revolution of Jupiter around the sun lies outside the reach of any moral categories as something that is wholly insusceptible to any pertinent moral assessment.² Thus, “It is not the case that the orbiting of Jupiter around the sun is morally impermissible” does not entail “The orbiting of Jupiter around the sun is morally permissible.” Both “It is not the case that the orbiting of Jupiter around the sun is morally impermissible” and “It is not the case that the orbiting of Jupiter around the sun is morally permissible” are true. Although the predicates “morally permissible” and “morally impermissible” are contradictories over the domain in which they are applicable, they are both inapplicable outside that domain.

The example in the preceding paragraph has pertained to a process or a sequence of events, but the point of the example naturally extends to states of affairs. Suppose that we inquire whether the strong-force interconnectedness of the protons and neutron(s) in some atom of helium is morally permissible or morally impermissible. As before, the appropriate answer to such a question is that neither moral permissibility nor moral impermissibility can ever appositely be predicated of such a state of affairs. To ascribe either of those properties to such a state of affairs is to commit a conceptual error as well as an ethical error.

Thus, if error theorists want to evade Dworkin’s trilemmatic challenge, they will have to maintain that the domain over which the “x” variable ranges in the formulation above is empty. They will have to establish that, contrary to what most people are inclined to think, human conduct and attitudes are relevantly similar to the revolution of Jupiter around the sun and to the links among the subatomic particles in a helium atom. Neither moral impermissibility nor moral permissibility can ever appositely be predicated of human conduct any more than of Jupiter’s orbiting. Such is the claim which the error theorists have to sustain.

In two respects, however, any such rejoinder to Dworkin by error theorists will founder. First, such a riposte will lack any adequate motivation or basis. Worries about the queerness or incoherence of objectively prescriptive properties cannot serve as the motivation, since—as has been stressed—moral permissibility is not an objectively prescriptive property. An imputation of such permissibility to some decision or mode of conduct does not in itself affirm or presuppose the existence of anything that is endowed with to-be-pursuedness or not-to-be-doneness. Yet, given that the worries about objective prescriptiveness are otiose as the motivation for a radically revisionary elimination of the ethical domain, the rejoinder by the error theorists that has been broached above is without any rationale.³

Second, even if the error theorists could manage to adduce some rationale for that rejoinder, they would be taking a position that teems with disconcerting moral implications.⁴ Indeed, Pigden concedes as much in his modification and defense of the error theory. He allows that “though the negation of a moral proposition is not always a moral proposition, it seems to me that sometimes and in some contexts it is.” Under his version of the error theory—a version that is pretty much along the lines envisaged in the riposte to Dworkin which I have posited above—“some items that might reasonably be described as moral judgments will come out true.” (Pigden 2007: 451, emphases in original) Pigden offers the example “It is not the case that we ought to keep our marriage vows,” but he could have offered any number of examples that are considerably more disturbing. For instance, according to the error theory, “It is not the case that the Khmer Rouge acted wrongly when they tortured and murdered hundreds of thousands of their Cambodian countrymen” and “It is not the case that a man who rapes and tortures and murders a woman has thereby done something morally wrong” and “It is not the case that members of the Ku Klux Klan are morally obligated to refrain from lynching African-Americans” are straightforwardly true. Even if it were somehow true that there is no domain in which “not morally impermissible” entails “morally permissible,” the former predicate obviously entails itself. Thus, the proponents of the error theory have saddled themselves with an array of decidedly unpalatable conclusions on matters of morality, even if they forbear from any affirmative uses of moral predicates in framing those conclusions.

A Further Argumentative Strategy

We should briefly examine four other argumentative strategies that might be adopted by an error theorist. In the first place, such a theorist might contend that affirmative logically atomic moral judgments are neither true nor false. (Joyce 2001: 6–9) Holding that the moral properties ascribed by such judgments are necessarily uninstantiated, an error theorist might conclude that any ascriptions of those properties involve presuppositional failures and that such ascriptions therefore lack determinate truth-values. However, when that line of thought is applied to ascriptions of permissibility, it is plagued by at least two major problems.

For one thing, the problem of the absence of any adequate basis or rationale—a problem posed in Sect. 1.2 above—is unresolved by this new argumentative strategy. Even if Mackie’s worries about the queerness of objective prescriptivity are accepted arguendo, ascriptions of moral permissibility do not attribute any objectively prescriptive properties to anything. Consequently, Mackie’s worries do not offer any support to the assumption that the property of moral permissibility is necessarily uninstantiated. A radically revisionary assumption of that sort can scarcely prevail in the absence of any support.

In addition, Olson aptly rebuts the notion that ascriptions of uninstantiated properties are not false: “In general, I take claims that predicate non-instantiated properties of some individual or individuals to be false. For instance, a claim to the effect that some person is a witch (where being a witch involves being a woman with magical powers) is false. The same goes [in the eyes of an error theorist] for a claim to the effect that acts of torture are morally wrong.” (Olson 2014: 12–13, footnote omitted) As Olson suggests, the proposition “Acts of interrogational torture are not morally wrong” is true by the reckoning of an error theorist, whereas the proposition “The present King of France is not bald” is neither true nor false by the reckoning of a Strawsonian analysis of presuppositional failures. Consequently, whereas the proposition “The present King of France is bald” involves a presuppositional failure that deprives it of truth-aptitude, the proposition “Acts of interrogational torture are morally wrong” involves not a presuppositional failure of that kind but instead—by the reckoning of the error theorists—a straightforward fallacy.

Another Argumentative Strategy

Let us, then, ponder Olson’s own preferred argumentative tack. A lengthy quotation from his discussion is advisable here:

A better way out is to deny that the implications from ‘not wrong’ to ‘permissible’ and from ‘not permissible’ to ‘wrong’ are conceptual and maintain instead that they are instances of generalized conversational implicature. To illustrate, ‘not wrong’ conversationally implicates ‘permissible’ because normally when we claim that something is not wrong we speak from within a system of moral norms, or moral standard for short. According to most moral standards, any action that is not wrong according to that standard is permissible according to that standard. General compliance with Gricean maxims that bid us to make our statements relevant and not overly informative ensures that we do not normally state explicitly that we speak from within some moral standard when we claim that something is not wrong. But the implicature from ‘not wrong’ to ‘permissible’ is cancellable. The error theorist can declare that torture is not wrong and go on to signal that she is not speaking from within a moral standard. She might say something like the following: ‘Torture is not wrong. But neither is it permissible. There are no moral properties and facts and consequently no action has moral status.’ This would cancel the implicature from ‘not wrong’ to ‘permissible’. (Analogous reasoning of course demonstrates why the error theorist’s claim that torture is not morally permissible does not commit her to the view that torture is morally impermissible and hence morally wrong.) (Olson 2014: 14, footnotes omitted)

The chief problem that afflicts this line of reasoning is the same as the chief problem that afflicts each of the other argumentative strategies which we have examined. That is, the cancellation of the implicature from “not wrong” to “permissible” is without any error-theoretical basis or rationale. Mackie’s worries about the queerness of objective prescriptivity are no such basis, because—in contrast with moral properties such as badness and goodness and obligatoriness and wrongness—permissibility is not objectively prescriptive. An ascription of permissibility to some course of conduct or state of affairs does not attribute either to-be-pursuedness or not-to-be-doneness to anything. Thus, even when the claims of Mackie’s error theory are accepted arguendo, any cancellation of the implicature from “not wrong” to “permissible” on the basis of that theory is completely unfounded.

In the parenthetical sentence at the end of the quotation above, Olson refers to the cancellation of the implicature from “not morally permissible” to “morally impermissible.” If the claims of Mackie’s error theory are granted arguendo, the cancellation of this latter implicature does have a basis in his concerns about the queerness of objective prescriptivity. Hence, contrary to what the parenthetical sentence suggests, there is no error-theoretical symmetry between the cancellation of the implicature from “not morally wrong” to “morally permissible” and the cancellation of the implicature from “not morally permissible” to “morally wrong.” Only the latter cancellation has any grounding in the tenets of Mackie’s error theory.

A Further Argumentative Strategy

An error theorist might next contest my repeated assertion that moral permissibility is not an objectively prescriptive property. Such a theorist might contend that an ascription of moral permissibility to some mode of conduct MC entails the proposition “It is morally wrong (at least prima facie) to prevent someone from doing MC” and the proposition “It is morally wrong (at least prima facie) to punish someone for doing MC.” In other words, an error theorist might submit that an ascription of moral permissibility entails an ascription of not-to-be-preventedness and an ascription of not-to-be-punishedness.

The First Putative Entailment: A Hohfeldian Corrective

In fact, however, an ascription of moral permissibility carries neither of the entailments just mentioned. The fact that the first entailment does not obtain is evident from the work of the American jurist Wesley Hohfeld, whose disambiguation of the notion of rights has been hugely influential in moral and legal and political philosophy.⁵ When some person X has in relation to some other person Y a liberty-right to φ, X does not owe Y any duty to abstain from φ-ing. However, unless X also holds a claim-right against Y’s prevention of her from φ-ing, the prevention is not even prima facie wrong. There is no entailment between X’s holding of a liberty-right and X’s holding of a claim-right. Thus, for example, suppose that Mark and Luke are runners in a two-man race. Mark is morally at liberty to outrun Luke, and Luke is morally at liberty to outrun Mark. Each of them has a liberty-right to defeat the other in the race, and therefore neither of them has a claim-right against being defeated by the other. If Mark does outrun Luke, he will have prevented Luke from outrunning him. Conversely, if Luke outruns Mark, he will have prevented Mark from outrunning him. Each of the two men is permitted to prevent the other from performing an act-type—namely, winning the race—which each of them is permitted to perform.⁶ This situation and countless other structurally similar situations are overlooked by any error theorist who postulates a relationship of entailment between “is permitted to φ” and “cannot legitimately be prevented from φ-ing.”

The Second Putative Entailment

Is an error theorist on any firmer ground in appealing to an entailment between “is permitted to φ” and “cannot legitimately be punished for φ-ing”? At least two difficulties fatally undermine such a move. In the first place, if there were any such entailment, it would be ethical rather than logical. Specifically, it would derive from a retributivist conception of punishment. Under any of the main consequentialist conceptions of punishment, there can be circumstances in which the infliction of some punishment on an innocent person is not even prima facie wrong.⁷ Hence, error theorists who try to ground their doctrine on a relationship of entailment between “is permitted to φ” and “cannot legitimately be punished for φ-ing” are invoking an affirmative logically atomic ethical proposition—the retributivistic proposition “Punishing a person for acting permissibly is always wrong (at least prima facie).” They are invoking such a proposition to support their contentions that every affirmative logically atomic ethical proposition is false.

Furthermore, even a retributivist can and should accept that—for some types of punishment—there is in fact no entailment between “is permitted to φ” and “cannot legitimately be punished for φ-ing.” For example, suppose that a government permissibly reaches some decision on the location of a public-works project or on the saving of one endangered species at the expense of another. Disgruntled voters (whether acting on the basis of self-interest or on the basis of nonprudential concerns) will be acting permissibly if they switch their votes from the ruling party to the rival party at the next election, yet the voters are thereby permissibly punishing the government for the permissible decision that it reached. Their punishing of the government for permissibly adopting its policy is not even prima facie wrong.

A Final Argumentative Strategy

Finally, adverting to the fact that the concepts of permissibility and obligatoriness and forbiddenness are interdefinable, an error theorist might hope to show that ascriptions of permissibility always presuppose or entail ascriptions of obligatoriness. Suppose that every permissible course of conduct available to any person P is specified, and suppose that the specifications are collected into a vast disjunction. Let us designate as “Choice” the act of selecting among the disjuncts—regardless of what the selection turns out to be. In that event, Choice is the sole permissible course of conduct available to P. (Some philosophers would be inclined to characterize the vast disjunction as itself a disjunctive course of conduct, and would maintain that that disjunctive course of conduct is the sole permissible course of conduct available to P. My terse discussion here could easily be reformulated in that manner.)

Now, an error theorist might submit that “the sole permissible” entails “obligatory.” If Choice is the sole permissible course of conduct available to P, then any contrary course of conduct is not permissible. Consequently, the error theorist might contend, Choice is a morally obligatory course of conduct for P. Accordingly, such a theorist might insist, the error theory could not be committed to ascriptions of moral permissibility without also being committed to ascriptions of moral obligatoriness. Since the latter ascriptions are untenable if Mackie’s arguments about the queerness of objectively prescriptive properties are correct, so too the former ascriptions (which presuppose the latter ascriptions) are untenable. And so the error theory escapes my critique.

The fatal flaw in any such rejoinder by a defender of the error theory is that “the sole permissible” entails “obligatory” only if at least one course of conduct other than the sole permissible course of conduct is available. Since any other such course of conduct would be impermissible, the sole permissible course of conduct would be obligatory as the exclusive way of avoiding every course of conduct that is impermissible. However, my claim about the error theory is that it entails the proposition that absolutely every course of conduct is morally permissible. For the error theory, there are no courses of conduct other than those specified in the vast disjunction envisaged above. Hence, it is not the case that Choice is the exclusive way of avoiding any course of conduct that is morally impermissible—because, by the reckoning of the error theory, there are no morally impermissible courses of conduct to be avoided. Choice does not perform any wrongness-averting role. Therefore, although Choice is the sole morally permissible course of conduct open to P, it is not morally obligatory. In sum, because the lone reason for deeming the sole permissible course of conduct in any context to be obligatory is that that course of conduct is the exclusive way of eschewing every impermissible course of conduct, and because that lone reason is inapplicable in a discussion of the error theory and its insistence that there are no morally impermissible courses of conduct, the fact that the error theory has to regard every possible course of conduct as morally permissible does not carry with it any commitment to regarding Choice as morally obligatory. Ergo, while fully recognizing that the error theory excludes all ascriptions of moral obligatoriness, I can rightly maintain that that theory entails the proposition that every possible course of conduct is morally permissible.

At this point, a defender of the error theory might seek to have recourse in a slightly different fashion to the interdefinability of permissibility and obligatoriness and forbiddenness. Such a defender might assert that, in light of the interdefinability of those concepts, the queerness or incoherence afflicting the notions of obligatoriness and forbiddenness must likewise afflict the notion of permissibility. If the propositions “Some course of conduct CX is morally obligatory” and “Some course of conduct CY is morally forbidden” are both false because their truth depends on the existence of properties that are irredeemably queer, then the truth of “Some course of conduct CZ is morally permissible” must similarly depend on the existence of such properties. So might argue an error theorist, who would submit that an ascription of permissibility to CZ presupposes the existence of queer properties even while denying that CZ partakes of them. Were there no such presupposition, the concept of permissibility would not be interdefinable with the concepts of obligatoriness and forbiddenness.

This additional recourse by a proponent of the error theory to the interdefinability of the key deontic concepts is misconceived, for that interdefinability does not involve any transfer of existential commitments from obligatoriness-ascribing propositions and forbiddenness-ascribing propositions to permissibility-ascribing propositions. When someone attributes permissibility to CZ, she thereby holds that CZ is not forbidden and that abstaining from CZ is not obligatory. Because the interdefinability of the concept of permissibility with the concepts of obligatoriness and forbiddenness is effected through negation in each case, any existential commitments carried by ascriptions of obligatoriness and forbiddenness are not similarly carried by ascriptions of permissibility. Indeed, if an ascription of permissibility is universally quantified across all possible courses of conduct—a universal quantification entailed by the error theory—it presupposes the inexistence of to-be-pursuedness and not-to-be-doneness. That presupposition is exactly what is implied by the interdefinability of permissibility and obligatoriness and forbiddenness. After all, a universally quantified ascription of moral permissibility maintains that nothing is ever morally forbidden and that nothing is ever morally obligatory. Consequently, such an ascription must presuppose the negation of any commitment to the existence of objectively prescriptive properties that is carried by an assertion that at least one course of conduct is morally obligatory or by an assertion that at least one course of conduct is morally forbidden. Accordingly, the interdefinability of permissibility and obligatoriness and forbiddenness is entirely consistent with my critique of the error theory as a doctrine that is committed to imputing moral permissibility to every possible course of conduct.

Dworkin and Smith on Status Skepticism

In the second half of his article, Smith concentrates on a passage in which Dworkin responded to those whom he designated as “status skeptics”—namely, philosophers who deny that moral utterances are ever truth-apt. Instead of joining the error theorists in maintaining that moral claims are always false, the status skeptics insist that such claims are always neither true nor false. In the draft of (Dworkin 2009) from which Smith’s article quotes, Dworkin replied to the status skeptics in part as follows:

I said that a status skeptic must find a way to reject the thesis he opposes, which is that moral judgments are candidates for objective truth, without also rejecting the first-order, substantive moral declarations he wishes to leave standing. I described two strategies he might use. He might claim, first, that the thesis he rejects is a second-order, philosophical claim different in meaning from the first-order substantive judgments he does not mean to oppose….The strategy fails unless he can identify a claim about the status of moral judgments that he rejects and that satisfies the two conditions of pertinence and independence. He cannot do this. I tried to illustrate his difficulty in the previous discussion: I canvassed many familiar arguments to suggest a more general thesis. Hume’s principle makes the strategy fruitless because a skeptical judgment about morality is still, in the pertinent sense, a moral judgment. (Dworkin 2009: 53, quoted in Smith 2010: 517)

As Smith remarks, and as is tersely indicated in this quoted passage, Dworkin held that status skeptics have to satisfy two requirements (which he labeled as “pertinence” and “independence”). First, they have to train their fire on a claim which they aim to deny. That is, they have to train their fire on an ascription of truth or truth-aptitude to some substantive moral judgment(s). Second, while contending that such an ascription is false, they have to refrain from contending that any substantive moral judgments are false or true. Dworkin argued that the status skeptics cannot satisfy both of these requirements, because there are no propositions which ascribe truth or truth-aptitude to moral judgments and which are not themselves moral judgments.

Smith responds to Dworkin’s arguments by submitting that—partly as a result of the strictures in Dworkin (1996)—nearly all status skeptics now accept that moral judgments are truth-apt and that such judgments express beliefs. However, Smith maintains that status skeptics think that the beliefs expressed through moral judgments are distinctive in being fully constituted by desires (or desire-like states) concerning non-moral matters of fact. Accordingly, the truths which those judgments articulate are also distinctive in being the contents of beliefs that are wholly constituted by desires (or desire-like states). Smith further declares that these claims propounded by status skeptics are not moral contentions but are instead purely philosophical in character. They are theses about the metaphysics of moral beliefs. Consequently, Smith proclaims, the status skeptics have met Dworkin’s challenge. Though they no longer deny that moral judgments are truth-apt, they do deny that the beliefs expressed by such judgments are metaphysically on a par with the beliefs expressed by assertions about non-normative and non-evaluative matters of fact. (Smith 2010: 518–520)

In the published version of Justice for Hedgehogs, Dworkin replied to Smith rather quizzically in a lengthy endnote. (Dworkin 2011: 432–433 n20) Although I concur with the gist of Dworkin’s remarks in that endnote and in the relevant portions of the main text of his book, I shall not respond directly to Smith here—for two reasons. First, the rest of this paper expounds a radically different understanding of the project pursued by some of the philosophers whom Dworkin designated as “status skeptics.” As will become apparent, my exposition of that alternative understanding can serve as an indirect reply to Smith.

Second, in his rejoinder to Dworkin on behalf of status skeptics, Smith explicitly takes for granted (at least arguendo) the truth of a doctrine that is often known as “motivational internalism” or “judgment-internalism.” According to the proponents of that doctrine, anyone who sincerely harbors some moral conviction is thereby vested with a disposition to conform his or her conduct to the content of that conviction. Although a particular disposition of that sort might be exceeded in strength by some countervailing inclination in a given context, it exists even in such a context where it goes unmanifested. Judgment-internalists believe that the existence of such behavioral dispositions in people who harbor moral convictions is a conceptual matter; in the eyes of these philosophers it is conceptually true that, if somebody is not at all disposed to refrain from φ-ing, he or she does not genuinely harbor the conviction that φ-ing is morally wrong. Elsewhere, I have argued at length that judgment-internalism elaborated as a conceptual thesis is fallacious.⁸ Instead of marshaling those arguments afresh here, I will simply repeat that Smith’s retort to Dworkin on behalf of status skeptics does not even get off the ground without the doctrine of judgment-internalism.

Before this paper presents its reconception of expressivism, however, we should note one feature of Smith’s discussion that also appears in many other discussions of these matters. Smith presumes that, because status skepticism as he elaborates it is consistent with a multitude of relatively concrete moral judgments such as “Lying is morally wrong,” it is not itself a moral thesis and is instead a purely philosophical thesis. (Smith 2010: 519) Such an inference is based on a misunderstanding of Dworkin’s anti-Archimedeanism and of cognate positions such as my insistence that moral realism is a moral doctrine. As Dworkin emphasized and as I have elsewhere emphasized (Kramer 2009: 3–5 et passim), meta-ethical theses are typically pitched at very high levels of abstraction and are therefore typically consistent with wide arrays of relatively concrete moral judgments. In that respect, meta-ethical theses are like other highly abstract moral doctrines. Given as much, the fact that status skepticism as Smith elaborates it is consistent with “Lying is morally wrong”—and with a myriad of other relatively concrete moral judgments—is not a sufficient basis for the conclusion that status skepticism is a non-moral thesis. Such skepticism as limned by Smith is a doctrine which holds that moral beliefs are derivatives or shadows of conative attitudes. As has already been stated, it is a doctrine that seeks to reconcile the nature of such beliefs with judgment-internalism. Now, although an exploration here of the character of judgment-internalism as a highly abstract moral position would take us too far afield, I have sought to provide such an exploration in my main writings on the topic elsewhere (cited above). Notwithstanding that the abstractness of judgment-internalism will leave it compatible with a vast throng of relatively concrete moral judgments, it is a moral position with some very general moral implications. So I have argued elsewhere. (Of course, if I were seeking to respond fully here to Smith’s defense of the Archimedeanism of status skepticism, I would need to recapitulate my previous ruminations on judgment-internalism. For my purposes in this paper, however, the sketchy remarks in this paragraph are sufficient. Smith’s delineation of status skepticism is itself quite laconic.)

Expressivism as Pragmatics, Quasi-Realism as Realism

In the rest of this paper, I will pursue several interrelated objectives. I will differentiate more sharply between expressivism and quasi-realism than most other philosophers; I will maintain that expressivism and quasi-realism are compatible but severable; I will contend that Simon Blackburn’s expressivism is properly understood not as an account of the semantics of moral discourse, but as an account of the pragmatics thereof; I will emphasize that expressivism so understood is philosophical rather than social-scientific and aetiological, and that it is not a moral doctrine; I will submit that quasi-realism is not “quasi” at all, in that it is equivalent to moral realism as a moral doctrine; and I will indicate that moral realism as a moral doctrine calls for some reorienting of meta-ethical enquiries rather than for the discontinuation thereof.

As my use of the terms “expressivism” and “quasi-realism” suggests, the rest of this paper is in some respects a reconstruction of the theorizing in which Blackburn has engaged. However, my aims are not exegetical, and there will be some differences between Blackburn’s theorizing and my project here. For example, whereas Blackburn repeatedly characterizes his approach as a reconciliation of ethics with naturalism, I readily allow that moral realism as a moral doctrine carries certain non-naturalistic existential commitments—provided that those commitments are construed minimalistically. In any event, although there are some substantial portions of Blackburn’s writings (especially his later writings) that are well captured by my exposition, there are other portions (especially in his earlier writings) that cannot be accommodated herein. Thus, notwithstanding that much of what follows is in some respects a reconstruction of Blackburn’s ideas, and notwithstanding that I shall occasionally point out how my elaboration of expressivism can shed light on certain aspects of his writings, this paper should be judged on its philosophical merits rather than on its exegetical merits.

Worth noting also is that the rest of this paper might not have been to the liking of Dworkin if he had lived to see it. For one thing, although there are many affinities between my elaboration of moral realism as a moral doctrine and his anti-Archimedean critique of meta-ethics, there are also some significant differences.⁹ A few of those differences will become apparent in the course of this paper. Equally important in the present context is that Dworkin might not have been receptive to my commendation of expressivism as a philosophical account of the pragmatics of moral discourse. On the one hand, he happily accepted that social-scientific research on the aetiology and effects of the use of moral concepts in people’s interaction can be fully creditable as a scholarly endeavor. (Dworkin 2011: 67) On the other hand, I am somewhat more doubtful that he would have granted that a distinctively philosophical account of the functioning of moral discourse—an account that is not a moral justification—can be valuable. Given that I shall draw an analogy between Blackburnian expressivism and H.L.A. Hart’s philosophy of law, and given Dworkin’s wariness of Hart’s jurisprudential theorizing, I am not robustly confident that Dworkin would have found much value in expressivism as a philosophical elucidation of the pragmatics of moral discourse. Still, I am likewise not robustly confident that he would have frowned upon such an endeavor. In any event, this portion of my paper is not a defense of Dworkin any more than it is straightforwardly an exegesis of Blackburn.

Expressivism and the Pragmatics of Moral Discourse

Expressivism has traditionally been regarded as an exposition of the semantics of moral utterances. It has been espoused, with varying degrees of sophistication and subtlety, to assimilate such utterances to statements that lack any propositional content (for example, interjections). When expressivism is expounded along these lines—in accordance with what Dworkin had in mind when he labeled expressivism as “status skepticism”—its proponents need to come to grips with difficulties such as the Frege–Geach problem.¹⁰ Those difficulties arise from the discrepancy between the propositional grammar of numerous moral utterances and the supposedly non-propositional contents of such utterances.

Expressivism so oriented is misguided, partly because the contents of moral judgments are propositional and irreducibly normative—and because the propositionality and irreducible normativity of their contents are moral matters—and partly because expressivism can be pertinently developed instead as a philosophical account of the pragmatics of moral discourse. That is, instead of aiming to supply an exposition of what moral utterances mean, expressivism should be aiming to supply an exposition of what people do by engaging in such utterances and by articulating them in propositional forms. It should be endeavoring to chart what people achieve at practical levels by suffusing their interactions with moral judgments. Alternatively stated, expressivism should be endeavoring to recount the important social functions that are performed by the inclusion of such judgments in human intercourse.

A Jurisprudential Analogy

I believe that Blackburn, especially in his later work, has redirected expressivism along the lines just suggested. However, I shall first adduce here an analogy—from the domain of legal philosophy—to clarify and amplify the foregoing paragraph. In the fifth chapter of The Concept of Law, Hart sought to illuminate the nature of the legal norms that he (somewhat unhelpfully) styled as “secondary.” Secondary legal norms are of three main types: (1) norms that empower and obligate officials to ascertain the laws of their society authoritatively; (2) norms that empower officials and ordinary citizens to bring about alterations in the existing laws; and (3) norms that empower and obligate officials to ascertain authoritatively whether any of the existing laws have been contravened (and to deal appropriately with any contraventions). To prod his readers to reflect on the important social functions that are performed through the existence and operativeness of secondary norms, Hart imagined a situation in which people’s interaction is not structured by any such norms. As he argued, such a situation would be crippled by rampant uncertainty, gross inefficiency, disorderliness, and stagnancy. Social interaction on a large scale for an extended period of time could not credibly occur under such circumstances. By averting or greatly alleviating those problems, secondary norms perform immensely important roles in any society. (Hart 1994: chap. 5)

Philosophy rather than Social Science or Moral Justification

In three main respects, the discussion of secondary legal norms in Hart (1994) is philosophical rather than an exercise in social science or moral justification. First, although Hart has sometimes been accused of indulging in anthropological speculations from his armchair,¹¹ and although some of his inaptly chosen wording has encouraged such accusations, his project is best understood very differently. Hart did not embark upon some genealogical or aetiological enquiry at all. Rather, as my preceding paragraph has indicated, he was seeking to highlight the important functions performed by secondary norms. He did so by prescinding from all the effects of secondary norms and by then pondering what the patterns of intercourse among human beings would be like without those effects. His prescinding from those effects was an abstract thought-experiment, rather than the postulation of a society that ever has existed or ever could exist. By imagining the absence of secondary norms, he rightly presumed, we can vividly grasp the far-reaching import of such norms in every actual society. Hart undertook a philosophical quest for clarification, rather than an anthropological quest for origins or causes.

Just as we should not take Hart’s discussion to be genealogical or aetiological, so too we should not construe it as an effort to provide a moral justification of law or of legal institutions. Hart as a legal positivist was well aware that the effects of secondary norms can be malign as well as benign. Those effects are momentously far-reaching, but they are not always morally commendable. Indeed, they can be morally pernicious. Although the effectuation of secondary norms in a functional system of law is undoubtedly necessary for the sway of justice and freedom and security and prosperity in a society, it is likewise necessary for the introduction and sustainment of noxious institutions such as chattel slavery. Accordingly, Hart’s exposition of the role of secondary norms is not to be construed as an endeavor to reveal the inherent moral worthiness of that role.¹² It is instead a philosophical elucidation of some of the major institutions that structure the lives and enterprises of human beings in any society.

Closely related to what has just been said is a third respect in which Hart’s discussion of secondary norms is philosophical. At a philosophical level of abstraction, it delineates some necessary features of any credibly possible society that exists on a large scale for a sustained period. Although some of the secondary norms recounted by Hart could be absent in an utterly fanciful large-scale society that never has existed and never will exist, secondary norms of all three main types are present in every actual large-scale society. Their presence is necessary for the cohesion and flexibility that enable any such society to persist. In that sense, they are existence-conditions for any large-scale society.

Expressivism as an Exposition of the Functions of Moral Discourse

Expressivism as an account of the pragmatics of moral discourse is quite closely similar to Hart’s project in a number of respects, though of course expressivism focuses not on legal norms but instead on the invocation of moral concepts by people in their multifarious dealings with one another. Expressivism does not address itself exclusively to fine-grained matters of pragmatics such as Gricean implicatures. It addresses itself at least as much to more coarse-grained matters of what people are able to do when they employ rather than eschew moral categories in their daily intercourse. It explores the important and wide-ranging effects—the important and wide-ranging social functions—that occur through the use and salience of such categories in human activities. In particular, it explores the effects that are made possible through the communication of moral attitudes as propositional judgments rather than simply as exclamations or other non-propositional utterances.

Although this paper is not exegetical, it seems to me that Blackburn has reoriented expressivism largely along the lines suggested here. Whereas in his early work he undertook to develop a logic that would enable him to handle the Frege–Geach problem (Blackburn 1993: chap. 10), he tackles that problem much more briskly in Blackburn (1998) where he addresses it by focusing on what people can do—in the coordination of their communications and conduct, for example—through the propositional structuring of moral discourse.¹³ His expressivism is directed not to explicating the meanings of moral assertions but instead to charting the practical workings and consequences of the activity of making such assertions.

Philosophy Rather than Social Science or Moral Justification

In parallel with Hart’s investigation of secondary legal norms, expressivism as understood here is distinctively philosophical. First, expressivism is no more an exercise in social-scientific genealogy or aetiology than is Hart’s exposition. On the one hand, Blackburn is occasionally incautious in his phrasing in ways that can easily lead to misunderstandings. He writes, for example, that “the moral proposition is designed or invented or emerges naturally as the focus for our practical transactions.” (Blackburn 1998: 70) This wording tends to convey the impression that Blackburn is indulging in anthropological speculations from his armchair. (In addition, the wording somewhat blurs the distinction between moral propositions and the formulations of those propositions.) However, his overall discussion makes clear that Blackburn has not set forth on any genealogical-aetiological quest for origins or causes in his reflections on moral discourse. Instead, at a philosophical level of abstraction, he is venturing to elucidate the functions of such discourse and specifically the functions of its propositional character.

Second, like Hart’s account of law, Blackburnian expressivism is not an exercise in moral justification. Expressivism concentrates on the fact that people use moral concepts and exchange moral assertions in their multifaceted intercourse with one another. To take note of that fact or to plumb its many practical implications is not per se to engage in moral assertions oneself. Though the practical dimension of moral discourse which Blackburn illuminates is immensely important, it is not necessarily benign. Odiously misguided moral judgments, which might be widely endorsed within some community, can perform the practical roles which Blackburn ascribes to moral discourse. Albeit the inclusion of moral assertions in people’s interactions is a necessary condition for the decency of those interactions on any large scale, it is decidedly not sufficient. As Nietzsche maintained (rather overheatedly) in The Genealogy of Morals, the inclusion of such assertions in people’s communications with one another can be a vehicle for the exertion and reinforcement of domination or for the propagation of highly unsavory attitudes. Thus, to delineate the practical ramifications of the activity of moralizing is not in itself to strive to vindicate that activity.

Third, like Hart’s discussion of secondary laws, Blackburnian expressivism recounts some necessary conditions for the persistence of any credibly possible large-scale community (or even any credibly possible small-scale community) over an extended period. Because of the key practical roles of moral assertions in coordinating people’s conduct and attitudes, the occurrence of such assertions in people’s dealings with one another—in varying ways and to varying extents—is a necessary condition for their living in proximity with one another sustainedly. Although a large-scale society in which people never form and communicate moral judgments is a logical possibility, it is not a credible possibility; there never has been and never will be such a society. Accordingly, the general patterns of interaction outlined by Blackburn are existence-conditions for any large-scale society.¹⁴

Quasi-Realism is Moral Realism as a Moral Doctrine

Expressivism operates at the level of the statements exchanged in moral discourse, as it seeks to highlight what people do and are enabled to do through the issuance of such statements. It is an account of how moral assertions function in human affairs, rather than an account of what those assertions mean. Quasi-realism—the other half of Blackburn’s endeavors—operates, by contrast, at the level of the propositions expressed in moral discourse. In a minimalistic fashion and with a philosophical degree of abstraction, it aims to expound the meanings and truth-conditions and statuses of those propositions. Instead of addressing the pragmatics of moral assertions, it addresses their semantics. It also, of course, addresses the metaphysics of moral principles and properties in a minimalistic fashion.

Quasi-realism as I understand it is moral realism as a moral doctrine. That is, it affirms the objectivity of morality in several different senses, and it does so entirely on moral grounds. It addresses the semantics and metaphysics and epistemology of morality, but it comes to grips with all such issues as substantive moral matters. It underscores the mind-independence of all basic moral principles and of many derivative moral principles; it contends that there are determinately correct answers to a myriad of moral questions; it affirms the categorical prescriptiveness of moral principles and the other ways in which such principles are uniformly applicable; it highlights the invariance of morality in major respects, while also readily acknowledging the variability of morality in other respects; it accounts for the knowability of moral principles and for the transindividuality of our epistemic access to them; it expounds the ways in which moral deliberations can be impartial and thus truth-conducive; and it maintains that any number of moral assertions are truth-apt and that some of them are true.¹⁵ In keeping with its minimalist approach to truth and facts and reality and properties, quasi-realism upholds the objectivity of morality in all these dimensions on moral grounds. Since moral realism is appropriately understood as an insistence on the objectivity of morality in all these dimensions—rather than as a doctrine about arcane entities whose reality and nature are determined by non-ethical considerations—moral realism as a moral doctrine is moral realism tout court. And since quasi-realism is moral realism as a moral doctrine, it too is moral realism tout court. There is nothing quasi about quasi-realism.

The Distinctness of Expressivism and Quasi-Realism

Of cardinal importance here is to keep in view the distinctness of expressivism and quasi-realism. Those two doctrines go together very nicely, of course,¹⁶ but they are not to be conflated. They address different phenomena and pursue different aims and employ different techniques. Expressivism focuses on the sundry practices in which people form and communicate moral judgments. It seeks to distill the functions that are performed by the inclusion of such judgments in those practices as people pursue various purposes that are furthered through their participation in moral discourse. By contrast, quasi-realism—or moral realism as a moral doctrine—focuses on moral principles and properties as it expatiates on the nature and objectivity of those principles and properties. The questions which it addresses are all internal to the domain of ethics. Its answers to those questions are grounded on moral considerations, and carry moral implications (though, because those answers are typically pitched at very high levels of abstraction, they are typically consistent with a wide array of concrete moral positions).

I have already remarked upon one way in which these distinct projects are often run together: that is, they are run together when expressivism is taken to be a theory of the semantics of moral judgments. Expressivism elaborated along those lines is indeed a version of status skepticism that deserves the censure which Dworkin directed at it. However, Dworkin himself was just as guilty of obscuring the major differences between expressivism and quasi-realism when he denied the possibility of the former by observing that it can never successfully come to grips with the issues that are addressed by the latter. He wrote as follows: “Of course there are interesting questions of anthropology and of personal and social psychology that are second-order in the sense that they are about moral judgment but do not themselves call for moral judgment. But there are no distinctly philosophical questions of that kind; in particular the question whether moral judgments can be true or false is a substantive moral issue, not a distinct meta-ethical one.” (Dworkin 2011: 67)

The first sentence and the closing half of the second sentence in this quotation are uncontroversial in themselves. Blackburn would certainly agree with each of those portions. What bespeaks confusion on Dworkin’s part in this quotation is his assertion that “there are no distinctly philosophical questions of that kind.” Strictly speaking, that assertion is true; as I have already emphasized, the questions addressed by anthropologists and sociologists who study human beings’ practices of moral discourse are not the questions addressed by expressivists.¹⁷ However, although Dworkin’s assertion is true, it is extremely misleading. A distinctively philosophical effort to answer the questions tackled by social scientists would be misguided, but a distinctively philosophical effort to achieve the elucidation sought by expressivists is perfectly sensible. Dworkin attempted here to substantiate his contrary view by pointing out that questions about the truth-aptitude of moral judgments are substantive moral matters. In so arguing, he was guilty of an ignoratio elenchi. Although he was of course correct in stating that the truth-aptitude of moral judgments is a matter internal to the domain of morality, he was quite wrong to suggest that that substantive moral matter or any other substantive moral matter is what expressivists qua expressivists aim to address. His error on that point resulted from his elision of the distinction between expressivism and quasi-realism. Expressivism’s austerely philosophical methods are indeed inapposite for handling the issues on which quasi-realists concentrate, but expressivists qua expressivists are not grappling with those issues. They are addressing neither the aetiological questions that preoccupy anthropologists nor the ethical questions (the highly abstract ethical questions) that preoccupy quasi-realists. Once the distinctiveness of their focus is duly grasped, and once the large differences between expressivism and quasi-realism are therefore likewise duly grasped, Dworkin’s doubts about the possibility of expressivism prove to be unfounded. In particular, we can then see that nothing in expressivism implies that the matters addressed by quasi-realists are anything other than substantive ethical matters.

Worth noting briefly here is another shortcoming in the quotation from Dworkin. He appeared to think that the status of meta-ethical positions as meaningful second-order positions—a status which he believed to be illusory, of course—would differentiate them from substantive ethical doctrines, which are first-order. That way of distinguishing between meta-ethics and ethics is very common, but it is misconceived. Although I join Dworkin in contending that all meta-ethical doctrines are ethical doctrines, I readily accept that meta-ethical doctrines are second-order. Also second-order are countless ethical theses that are uncontroversially classifiable as substantive ethical theses. In a defense of moral realism as a moral doctrine, the key dichotomy is not between the first-order and the second-order; instead, it is between the internal and the external. A defender of moral realism as a moral doctrine maintains that all meta-ethical propositions—all propositions about the existence or nature or objectivity of moral principles and properties—are internal to the ethical domain rather than external.¹⁸ They are second-order theses, but, at high levels of abstraction, they are just as squarely within the ethical domain as are any other substantive ethical theses. In that respect they are parallel to meta-mathematical or meta-logical theses, which are second-order but squarely within the mathematical or logical domain. (When I claim that a thesis T is internal to the ethical domain, I am claiming that T is grounded on ethical considerations and that it carries some ethical implications. In other words, any factors that are sufficient to render T true are ethical considerations—or include ethical considerations—and T is inconsistent with some substantive ethical positions. In the senses of “austere” and “neutral” specified by Dworkin (1996: 92), a thesis internal to the ethical domain is neither austere nor neutral.)

The Importance of Minimalism

Of central importance to moral realism as a moral doctrine (or to Blackburn’s quasi-realism) is its minimalist account of truth and reality and facts and properties.¹⁹ Precisely because those notions and cognate notions are all to be construed minimalistically, debates over the existence or nature or objectivity of moral principles and properties are substantive ethical debates. When we ask whether acts of torturing babies for pleasure really are morally wrong, or when we ask whether the moral principle “Acts of torturing babies for pleasure are morally wrong” is true, or when we ask whether it is a fact that acts of torturing babies for pleasure are morally wrong, we are asking whether acts of torturing babies for pleasure are morally wrong. As Blackburn has colorfully written: “[W]e know [some] 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, if we like that kind of talk. We can add flowers without end.” (Blackburn 1998: 79) In sum, although any affirmation or denial of the reality of some moral property MP is a metaphysical thesis, it is pari passu a substantive moral thesis. An affirmation of the reality of MP maintains that some actual or possible objects of moral assessment—such as modes of conduct or outlooks or intentions or states of affairs—do or would partake of MP. A denial of the reality of MP maintains that no actual or possible objects of moral assessment ever partake of MP. Whether affirming or denying MP’s reality, we are in the domain of substantive morality.

Of course, as both Dworkin and Blackburn have emphasized, statements which ascribe reality to moral properties or truth to moral principles or factuality to moral propositions are not always pleonastic—just as a word-for-word repetition of a statement is not always pleonastic. Such ascriptions can amplify or clarify or intensify the moral judgments to which they are attached. They do so not by adding anything non-ethical to those judgments but instead by adding to their ethical substance. For example, when someone asserts that the mistreatment of a young boy by some older boys really is wrong, she might be intensifying the tenor of her censure to indicate that the mistreatment is heinous; or she might be indicating that the mistreatment is wrong irrespective of what the boys or anyone else may think; or she might be indicating that the matter is clear-cut in that there are no significant considerations which militate against the correctness of her verdict. These and many other ways of amplifying or clarifying or intensifying any ethical judgments are themselves ethical judgments. Hence, the upshot of a minimalist approach to truth and cognate notions is that appeals to truth or reality or objectivity in discussions of ethical matters are either reiterations of ethical judgments or further elaborations thereof.

Although the correspondence theory of truth is usually propounded as a rival to a minimalist account, minimalism is perfectly consistent with a deflationary version of the correspondence theory. With reference specifically to the domain of ethics, a deflationary version of the correspondence theory comprises two main theses:

• (1) A proposition which articulates the moral bearings of any mode of conduct in any given circumstances is true if and only if the proposition follows from the correct principles of morality that are applicable to the specified conduct in the specified circumstances.

• (2) A proposition which affirms the existence of some moral standard is true if and only if that standard is one of the correct principles of morality.

As was to be expected, this deflationary version of the correspondence theory presents moral truth or falsehood as a substantive moral property. To gauge the truth or falsity of a statement that recounts the moral bearings of some mode of conduct, we have to endeavor to discern the correct principles of morality that are applicable in the prevailing circumstances, and we have to engage in whatever moral ruminations are necessary to pin down the implications of those principles in those circumstances. To gauge the truth or falsity of a statement that affirms the existence of some moral standard, we have to enquire whether the standard is among the correct principles of morality. Again, any enquiry of that kind will be conducted through individual reflections and through exchanges of views among individuals.

In other words, if we let “p” stand for a moral proposition, the endeavors required for verifying or disconfirming “The proposition ‘p’ is true” are the same as those required for verifying or disconfirming “p” itself. Quite predictably, the epistemic corollaries associated with the deflationary version of the correspondence theory of truth are the same as those associated with the disquotational version of minimalism. Each account of truth generates the conclusion that there is no tenable distinction between ascertaining whether “p” is true and ascertaining whether p.

Moral Realism as a Moral Doctrine and the Place of Meta-Ethics

Dworkin often elicited antagonism from moral philosophers as he waged what seemed to be a thoroughgoing campaign against meta-ethics. Russ Shafer-Landau’s reaction is representative of the responses by quite a few other philosophers: “I found the first few chapters of Ronald Dworkin’s Justice for Hedgehogs very disconcerting….[T]wo distinct sources of worry kept pressing. My initial unease grew steadily to something approaching panic as I assimilated the underlying message of these chapters—metaethics is largely a sham; its central question—that of the status of ethical and moral views—is a pseudo-question. This was bad enough, surely, for someone who has spent the last dozen years devoted to trying to make progress on that question.” (Shafer-Landau 2010: 479) Such reactions are probably due to the supercilious tone of Dworkin’s prose as much as to the substance of his arguments. At any rate, whatever the pertinence of such reactions in the face of Dworkin’s anti-Archimedean onslaughts, they are not pertinent at all in response to my advocacy of moral realism as a moral doctrine. As should be evident from what has been said in this paper, and as I have emphasized in Kramer (2009), my espousal of moral realism as a moral doctrine is scarcely aimed at establishing that meta-ethics is a sham. In two respects, meta-ethics should persist and flourish.

First, if the field of meta-ethics is understood more expansively than I have construed it when I have proclaimed that all meta-ethical doctrines are substantive ethical doctrines, and specifically if that field is understood to include endeavors such as Blackburnian expressivism, then those endeavors should continue as they have unfolded hitherto. They are entirely consistent with moral realism as a moral doctrine, and they complement it nicely. Their status as distinctively philosophical enterprises is not under challenge at all in my writings.

Second, meta-ethics as I have generally heretofore understood it should also continue along its sundry paths—provided that the philosophers who proceed along those paths are cognizant that their metaphysical and semantic and epistemological enquiries are substantive ethical enquiries. To insist on the substantive ethical character of those enquiries is decidedly not to suggest that (in Shafer-Landau’s words) the question of “the status of ethical and moral views” is a pseudo-question. On the contrary, that topic is rightly one of the central foci of meta-ethics, and it is a topic squarely and sustainedly addressed by moral realism as a moral doctrine. Anyone who seeks to uphold the objectivity of ethics could hardly fail to address that question with the utmost earnestness.

Indeed, as I have mentioned earlier, moral realism as a moral doctrine is replete with existential commitments. Basic moral principles and many derivative moral principles such as “Torturing babies for pleasure is morally wrong” have always existed and will always exist. They are binding always and everywhere in all possible worlds; they would have existed even if no human beings or other rational moral agents had ever existed. (Kramer 2009: 26–48, 154–161) These existential commitments are fully consistent with my claim that moral realism as an affirmation of the objectivity of morality is a moral doctrine, for all of the foregoing commitments are to be construed minimalistically. When I declare that the principle “Torturing babies for pleasure is morally wrong” has always existed and will always exist, I am declaring that torturing babies for pleasure always has been morally wrong and always will be morally wrong. Such existential commitments are moral theses that have to be substantiated (or contested) on moral grounds.

As should be apparent, the label “quietism” is badly misguided when applied to moral realism as a moral doctrine.²⁰ As John McDowell has observed (McDowell 2009: 370–372), that label is typically used derogatorily to suggest that certain philosophers have—perhaps inadvertently—revealed the need for philosophical analyses which they decline to carry out. Whether because of laziness or incompetence or misplaced scruples, these philosophers have been remiss. So the allegation typically runs, when the term “quietism” is employed by critics.

Disparagement of this kind occurs, for example, in Ralph Wedgwood’s powerful book on the nature of normativity. (Wedgwood 2007) Wedgwood excoriatingly distances himself from quietism, which he characterizes as the view that “no substantive, non-trivial ‘account’ of normative thought or discourse is possible.” Exactly why he has placed cautionary quotation marks on the term “account” is not clear, but he expands on his conception of quietism as follows: “According to this quietist approach, nothing more can be said about the meaning of statements of the form ‘A ought to φ’ except things of the following forms: that this statement means that A ought to φ, that it can be used to state that A ought to φ, and to express the belief that A ought to φ, and so on.” (Wedgwood 2007: 18, italics in original) Wedgwood disdainfully retorts to the quietist approach which he has sketched: “[I]t should be clear that [the quietist approach] is hardly obviously true. Indeed, it would seem to me incredible that it could be an absolutely unanalysable feature of a particular thought or statement that it is about one thing rather than another….It is obviously incredible to suppose that the name ‘Socrates’ refers to a particular individual purely by magic, as it were.” (Wedgwood 2007: 19, emphasis in original)

In two respects, Wedgwood’s derision is unfounded. First, as I have remarked in my far longer discussion of minimalism in Moral Realism as a Moral Doctrine (Kramer 2009: 265), the appearance of triviality in minimalist approaches to truth and meaning and cognate notions is only an appearance. Those approaches have to be defended, and have indeed been defended elsewhere, through complex argumentation on a number of points. Second, and perhaps even more important, I have already emphasized above that my espousal of moral realism as a moral doctrine does not involve any calls for the discontinuation of meta-ethical philosophizing. To highlight the substantive ethical character of meta-ethical theses is hardly to imply that such theses are in no need of substantiation or explication. When one endeavors to support or impugn second-order ethical theses at high levels of abstraction, just as when one endeavors to support or impugn other ethical theses, one has to proceed with arguments rather than with silence.

A variant of this second response to Wedgwood is pertinent as a rejoinder to James Dreier, who presents the following two formulations (Dreier 2004: 35):

• (E) Edith said that abortion is wrong.

• (G) Its being the case that (E) consists of nothing more than ____.

Dreier chidingly refers to “lazy theorists [who] might just try [to fill in the blank by] more or less repeating (E).”²¹ Filling in the blank by largely repeating E would of course be correct, but no proponent of moral realism as a moral doctrine has to rest content with that stark approach. Equally apt as a way of filling in the blank would be “Edith’s having ascribed to the act-type of abortion the basic moral status that is conferred or would be conferred on that act-type by any moral principle that prohibits abortion.” This latter way of filling in the blank is more controversial than the E-repeating approach—it would need to be defended (on moral grounds) against devotees of moral particularism, for example—but a proponent of moral realism as a moral doctrine can happily adopt it.

A Pithy Conclusion

While moving briskly, this paper has sought in several ways to clarify and defend moral realism as a moral doctrine. It has begun by upholding Dworkin’s anti-Archimedean critique of the error theory against some of Smith’s strictures, and has then briefly suggested how a proponent of moral realism as a moral doctrine would respond to Smith’s defense of the Archimedeanism of status skepticism. Thereafter, this paper has moved to its chief endeavor. By differentiating clearly between expressivism and quasi-realism (or moral realism as a moral doctrine), I have highlighted both their distinctness and their compatibility. In so doing, I have sought to underscore the affinities between Blackburnian quasi-realism and moral realism as a moral doctrine. Finally, this paper has contended—in line with my earlier work on these matters—that moral realism as a moral doctrine points to the need for some reorienting of meta-ethical enquiries rather than for the abandoning of them. The tone of haughty dismissiveness in some of Dworkin’s discussions of meta-ethics is an irksome distraction from the sorts of adjustments that will be salutary.²²