Richard Joyce

To reject a false theory on the basis of an unsound argument is, in my opinion, as much an intellectual sin as to embrace a false theory. Thus, although I am no fan of any particular form of moral rationalism—and, indeed, on occasion have gone out of my way to criticize it—when rationalism is assailed for faulty reasons I find myself in the curious position of leaping to its defense (which goes to show that in philosophy it isn’t the case that one’s enemy’s enemy is one’s friend). This puts me at something of a dialectical disadvantage, since my “defense” of moral rationalism has strict limits: I will defend it against specific kinds of criticism, but I have no interest in defending it simpliciter. It is important to bear in mind that what is in dispute between Shaun Nichols and myself is not the truth or falsity of moral rationalism, but rather what kind of evidence bears on the matter.

The popular expedient of identifying noncognitivism with the claim that moral judgments are neither true nor false leaves open the question of what kind of thing a moral judgment is—an indeterminacy that has led to decades of confusion as to what the noncognitivist is more precisely committed to. Sometimes noncognitivism is presented as a claim about mental states (“Moral judgments are not beliefs”), sometimes as a claim about meaning (“X is morally good” means no more than “X: hurray!”), sometimes as a claim about speech acts (“Moral judgments are not assertions”). Focus on the last two possibilities. The former calls for a translation schema from a propositional surface grammar to a non-propositional deep structure. Such schemata from the noncognitivist are familiar to students of metaethics. (Cf. A.J. Ayer’s claim that in saying “You acted wrongly in stealing that money” one is “not saying anything more than … ‘You stole that money,’ [but] in a peculiar tone of horror.”) It is less widely realized that the noncognitivist is not obliged to offer any such translation schema, for she might instead plump for the last option, of formulating noncognitivism as a theory not of meaning but of use. Perhaps the moral cognitivist is correct about the meaning of moral sentences (there is a wide range of possibilities here) but wrong about the way people use moral sentences: perhaps people do not assert moral sentences, perhaps the nature of acceptance of a moral claim is not belief.

The lead text of this book is based on primatologist Frans de Waal’s 2003 Tanner Lectures at Princeton University, to which he adds three short appendices. There are commentaries by Robert Wright, Christine Korsgaard, Philip Kitcher, and Peter Singer, followed by a 20-page response. Josiah Ober and Stephen Macedo provide a brief introduction. As befits a Tanner lecturer, de Waal’s scope is broad, his writing accessible, and the pace lively. He continues his crusade against the “veneer theory”—the idea that humans are naturally sociopaths and that empathy, altruism, and morality are the result of recent cultural forces. The extent to which de Waal dichotomizes this debate, however, masks much of interest. One can agree with the seemingly obvious fact that humans are obligatorily gregarious organisms—replete with various prosocial mechanisms—while still expressing reasonable doubt that our capacity to make moral judgments is the result of an innate mechanism dedicated to that task. De Waal is an unparalleled source of rich first-hand primatological data, arguing cogently that humans show continuity with other apes regarding empathy, consolation behavior, peacemaking, and reciprocal relations. Yet he admits these “building blocks” are “by no means sufficient” for morality, and in his Response discusses two further levels of morality: rule enforcement for the community’s good, and disinterested moral reasoning. He doesn’t address whether this third, uniquely human, level is a distinct innate adaptation, apparently not appreciating that if it isn’t, and yet is necessary for being literally accorded a “moral sense”, then moral nativism is false and at least some of human morality is “veneer.” In characteristic non-philosopher manner, de Waal ultimately rejects the question as “semantics” and thus “mostly a waste of time”. It seems a surprising blemish that the author of a book subtitled “How Morality Evolved” should grow impatient with those wondering what precisely is necessary and sufficient for morality, and what specifically has evolved.

“Nihilism” (from the Latin “nihil” meaning nothing) is not a well-defined term. One can be a nihilist about just about anything: A philosopher who does not believe in the existence of knowledge, for example, might be called an “epistemological nihilist”; an atheist might be called a “religious nihilist.” In the vicinity of ethics, one should take care to distinguish moral nihilism from political nihilism and from existential nihilism. These last two will be briefly discussed below, only with the aim of clarifying our topic: moral nihilism. Even restricting attention to “moral nihilism,” matters remain indeterminate. Its most prominent usage in the field of metaethics treats it as a synonym for “error theory,” therefore an entry that said only “Nihilism: see ERROR THEORY” would not be badly misleading. This would identify moral nihilism as the metaethical view that moral discourse consists of assertions that systematically fail to secure the truth. (See Mackie 1977; Joyce 2001.) A broader definition of “nihilism” would be “the view that there are no moral facts.” This is broader because it covers not only the error theory but also noncognitivism (see NONCOGNITIVISM). Both these theories deny that there are moral facts—the difference being that the error theorist thinks that in making moral judgments we try to state facts (but fail to do so, because there are no facts of the type in question), whereas the noncognitivist thinks that in making moral judgments we do not even try to state facts (because, for example, these judgments are really veiled commands or expressions of desire). (In characterizing noncognitivism in this way, I am sidelining various linguistic permissions that may be earned via the quasi-realist program (see QUASI-REALISM).) While it is not uncommon to see “nihilism” defined in this broader way, few contemporary noncognitivists think of themselves as “nihilists,” so it is reasonable to suspect that the extra breadth of the definition is often unintentional. Both these characterizations see moral nihilism as a purely metaethical thesis...n

John Mackie’s moral error theory is so closely associated in people’s minds with his arguments from relativity and from queerness that one might overlook the fact that there may be numerous other, and possibly better, ways of establishing that metaethical position. Perhaps, indeed, there are even further resources for arguing for a moral error theory to be unearthed in Mackie’s own book. I have in mind Mackie’s thesis of moral objectification: that the “objective prescriptivity” with which our moral judgments are imbued is the result of our “tendency to read our feelings into their objects” (1977: 42). Mackie invokes Hume’s famous projectivist image of the human mind’s “great propensity to spread itself on external objects,” and, indeed, it is in his book-length analysis of Hume’s moral theory (Mackie 1980) that the topic receives a more careful discussion than in Ethics: Inventing Right and Wrong. In both books he musters some considerations in favor of the thesis, and reveals to us that he thinks that “it is very largely correct” (1980: 72).

In his contribution to this volume, Paul Bloomfield analyzes and attempts to answer the question “Why is it bad to be bad?” I too will use this question as my point of departure; in particular I want to approach the matter from the perspective of a moral error theorist. This discussion will preface one of the principal topics of this paper: the relationship between morality and self-interest. Again, my main goal is to clarify what the moral error theorist might say on this subject. Against this background, the final portion of this paper will be a discussion of moral fictionalism, defending it from some objections.

This paper concerns the relation between two metaethical theses: moral naturalism and moral skepticism. It is important that we distinguish both from a couple of methodological principles with which they might be confused. Let us give the label “Cartesian skepticism” to the method of subjecting to doubt everything for which it is possible to do so—usually by introducing alternative hypotheses that are consistent with all available evidence (e.g., brains in vats). Let us give the label “global naturalism” to the principle that requires of any item which we admit into our ontology that it “fits” (in some manner or cluster of manners to be specified) with our naturalistic scientific worldview. One might be both a Cartesian skeptic and a global naturalist, if the latter principle is something that has survived the former test procedure. Alternatively, one might have adopted global naturalism for some other reason, while having little patience for the Cartesian method of doubt. Moral naturalism is the metaethical view that moral entities (e.g., properties like goodness and evil) fit within the scientific image of the world. The moral naturalist will probably be a global naturalist, but need not be: It is consistent with allowing non-natural entities into one’s ontology that one happens to think that moral properties are of the natural variety. Moral skepticism denies that moral entities fit within our scientific worldview. One way of denying moral naturalism is to be a moral error theorist: to hold that our moral discourse attempts to make reference to moral properties, but these properties do not exist.1 Another way of denying moral naturalism is to be a noncognitivist: to hold that our moral discourse was never really in the business of referring to moral facts or properties in the first place, and ipso facto such facts or properties are not naturalistic. In this paper, the label “moral skepticism” denotes the disjunction of these two theses. Neither the error theorist nor the noncognitivist must be committed to global naturalism, but usually will be; indeed, this commitment will often be a motivating factor of their metaethical views..

Were I not afraid of appearing too philosophical, I should remind my reader of that famous doctrine, supposed to be fully proved in modern times, “That tastes and colours, and all other sensible qualities, lie not in the bodies, but merely in the senses.” The case is the same with beauty and deformity, virtue and vice. This doctrine, however, takes off no more from the reality of the latter qualities, than from that of the former; nor need it give any umbrage either to critics or moralists. Though colours were allowed to lie only in the eye, would dyers or painters ever be less regarded or esteemed? There is a sufficient uniformity in the senses and feelings of mankind, to make all these qualities the objects of art and reasoning, and to have the greatest influence on life and manners. And as it is certain, that the discovery above-mentioned in natural philosophy, makes no alteration on action and conduct; why should a like discovery in moral philosophy make any alteration?

The first objective of this chapter is to clarify what might be meant by the claim that human morality is innate. The second is to argue that if human morality is indeed innate an explanation may be provided that does not resort to an appeal to group selection, but invokes only individual selection and so-called “reciprocal altruism” in particular. This second task is not motivated by any theoretical or methodological prejudice against group selection; I willingly concede that group selection is a legitimate evolutionary process, and that it may well have had the dominant hand in the evolution of human morality. There is a fact of the matter about which process, or which combination of processes, produced any given adaptation, and it is to be hoped that in time enough evidence might be brought into the light to settle such issues. At present, though, the evidence is insufficient regarding human morality. By preferring to focus on reciprocity rather than group selection I take myself simply to be outlining and advocating a coherent and uncomplicated hypothesis, which may then take its place alongside other hypotheses to face the tribunal of our best evidence.

Different versions of moral projectivism are delineated: minimal, metaphysical, nihilistic, and noncognitivist. Minimal projectivism (the focus of this paper) is the conjunction of two subtheses: (1) that we experience morality as an objective aspect of the world and (2) that this experience has its origin in an affective attitude (e.g., an emotion) rather than in perceptual faculties. Both are empirical claims and must be tested as such. This paper does not offer ideas on any specific test procedures, but rather undertakes the important preliminary task of clarifying the content of these subtheses (e.g., what is meant by "objective"? what is meant by "experience"?). Finally, attention is given to the relation between (a) acknowledging that the projectivist account might be true of a token moral judgment and (b) maintaining moral projectivism to be true as a general thesis

Facts about the evolutionary origins of morality may have some kind of undermining effect on morality, yet the arguments that advocate this view are varied not only in their strategies but in their conclusions. The most promising such argument is modest: it attempts to shift the burden of proof in the service of an epistemological conclusion. This paper principally focuses on two other debunking arguments. First, I outline the prospects of trying to establish an error theory on genealogical grounds. Second, I discuss how a debunking strategy can work even under the assumption that noncognitivism is true.

To hold an error theory about morality is to endorse a kind of radical moral skepticism—a skepticism analogous to atheism in the religious domain. The atheist thinks that religious utterances, such as “God loves you,” really are truth-evaluable assertions (as opposed to being veiled commands or expressions of hope, etc.), but that the world just doesn’t contain the items (e.g., God) necessary to render such assertions true. Similarly, the moral error theorist maintains that moral judgments are truth-evaluable assertions (thus contrasting with the noncognitivist), but that the world doesn’t contain the properties (e.g., moral goodness, evil, moral obligation) needed to render moral judgments true. In other words, moral discourse aims at the truth but systematically fails to secure it. If there is no such property as moral wrongness, for example, then no judgment of the form “X is morally wrong” will be true (where “X” denotes an actual action or state of affairs). Advocates of this position include Hinckfuss 1987; Joyce 2001; Mackie 1977 (see MACKIE, J. L.). Various forms of moral skepticism—some of which are arguably instances of the error theoretic stance—have been familiar to philosophers since ancient times. (See SKEPTICISM, MORAL.) Error theoretic views can be controversial—as in the case of religion and morality—or widely agreed upon—as in the case of ghosts and phlogiston. It is important to note that error theorists maintain that the judgments in question are erroneous not merely because of the absence of any objective moral facts sufficient to render them true, but also because of the absence of any non-objective moral facts sufficient to render them true. There is, for example, a kind of moral realist who maintains that moral properties are objective features of the universe (see REALISM, MORAL). There is also a family of metaethical views according to which moral properties are in some manner constituted by us—by our beliefs, attitudes, practices, etc..

Moral skepticism is the denial that there is any such thing as moral knowledge. Since the publication of The Myth of Morality in 2001, Richard Joyce has explored the terrain of moral skepticism and has been willing to advocate versions of this radical view. Joyce's attitude toward morality is analogous to an atheist's attitude toward religion: he claims that in making moral judgments speakers attempt to state truths but that the world isn't furnished with the properties and relations necessary to render such judgments true. Moral thinking probably emerged as a human adaptation, but one whose usefulness derived from its capacity to bolster social cohesion rather than its ability to track truths about the world. Essays in Moral Skepticism gathers together a dozen of Joyce's most significant papers from the last decade, following the developments in his ideas, presenting responses to critics, and charting his exploration of the complex landscape of modern moral skepticism.

Moral naturalism is the view that moral properties exist in a manner that fits with our scientific worldview. Might empirical discoveries about the genealogy of moral judgments (that, for example, they issue from an evolved psychological faculty) serve to undermine moral naturalism? One way of undermining moral naturalism is to show that moral properties do not exist at all. The possibility of genealogical considerations supporting this conclusion is examined and found to be weak. Alternatively, might empirical discoveries about the genealogy of moral judgments serve to vindicate some form of moral naturalism? This possibility is also explored and found to be unconvincing.