The call to abolish morality is rarely heard, but often rejected, which is not surprising because, on the face of it, abolishing morality seems like a horrible idea, one that would lead to massive suffering and disaster. But would it? Or is it possible that, as a few “abolitionists” have argued, morality is responsible for more suffering than it prevents? Difficult as it was, and remains, to abolish slavery, hunger, and torture, abolishing morality would be more difficult still. We can’t legislate it out of existence, eliminate its cause, or condemn morality as itself immoral. Morality is embedded in our language and in our vision of ourselves and our place in the world in a way that, we must hope, slavery, hunger, and torture are not. There is nothing that morality may not touch, and no side of any dispute that it cannot be called upon to support. But apart from its familiarity and its apparent usefulness, the main reason why most people would reject the call to stop speaking and acting as if there are moral truths about what is good, bad, right and wrong is that they believe that there are such truths. To convince someone to abolish or abandon morality, you must first get them to stop believing in moral properties, facts, and truths. You must get them to become moral anti-realists, but, as we shall see, even that is not enough.
The moral anti-realist has arrived at the idea that moral judgments are not true statements of objective facts; but moral anti-realism comes in two traditional flavors—non-cognitivism and the error theory. Both deny that there are such things as moral facts, but the absence of moral facts matters little to the non-cognitivist, who thinks that when we make moral judgments we aren’t even trying to state moral facts. Instead, we are expressing our feelings, emotions, attitudes, or stances, and/or attempting to influence the attitudes and behavior of others. A. J. Ayer startled a moralistic world with his claim that our “exhortations to moral virtue are not propositions at all, but ejaculations or commands which are designed to provoke the reader to action of a certain sort.” Non-cognitivists are usually quite happy to use moral language to express their attitudes and to manipulate others, so they are unlikely to be moral abolitionists.
By contrast, an error theorist claims that the notion of a moral fact makes enough sense for us to be able to say that there are no such things. Our “exhortations to moral virtue” either are propositions, or presuppose propositions, about what is objectively good, bad, right, or wrong. Here the absence of moral facts is very important because it is what makes each of these moral judgments false. This second way to develop moral anti-realism was first seriously defended by John Mackie.
In this paper I want to explore a dispute among error theorists about how to deal with what all sides admit are false moral judgments. The advice of the moral abolitionist is to give up moral language, and with it the arrogance and interference that a belief in the objectivity of morality often occasions. But the contrary advice of the moral fictionalist is to retain moral language and moral thinking because morality is too useful to be abandoned. My aim here is to clarify the choice that arises for one who has embraced a moral error theory, and to argue that moral abolitionism has much to recommend it. I will assemble some considerations that suggest that the death of moralizing might be good for the individual and for society, and that moral abolitionism may be more useful than moral fictionalism in helping us reach the goals most compassionate moralists seek.
1. Inventing Morality. In discussing John Mackie’s error theory, Simon Blackburn observed that since Mackie thought morality was infested with error, perhaps he should have abandoned moral concepts and ways of thought, or at least replaced them “by ones that serve our legitimate needs but avoid the mistake.” But Mackie was less inclined to abandon or replace moral thought and language than to make use of them. Like Hume, he believed that our habits of selfishness and confined generosity are incapable of generating sufficiently sociable behavior; and he concluded from this that “it is important that there should be a widespread tendency to act on moral grounds.” Important social practices like recognizing property and keeping agreements need to be supplemented by a “moral overlay” that allows us to say that the actions we favor are objectively required by objective circumstances and that some of the things we desire have a value of their own.
Unlike many moralists and moral philosophers who take the usefulness of moral beliefs for granted, Mackie directly addressed the question of whether morality does more harm than good. In Hume’s Moral Theory he defended his belief in the need for morality against those who recommend that we abandon or replace morality. He argued that if we supplement those important social practices like recognizing property and keeping agreements with the moral overlay, then people will be more likely to leave our possessions alone and to abide by their words.
Mackie put forward a second reason for preserving the moral overlay. He said that when people believe in an objective truth about what is right and wrong it is easier to support and rationalize legal decisions and sanctions. If we could not anchor the use of force in some claim to a legitimate, objective, and moral ground, that use of force and the useful practices it supports could be more easily challenged.
There may be something to this, but these two reasons for adopting the moral overlay are surprisingly weak. Consider the first. We all want others to respect our property, to tell the truth, and to keep their agreements, and it may be that when we insist that some practices are required by morality some people will be encouraged to adhere more closely to them. But it is not true that a practice plus a moral overlay will always be as fully, regularly, or willingly followed as that practice supplemented by some of the many available non-moral devices. We might promote the practice of keeping promises by early and extensive training in empathy, by strict surveillance and strong penalties for promise-breakers, or by massive doses of advertising by celebrities. Since we have access to these and other powerful ways to encourage promise-keeping, the abolitionist can argue that the moral overlay may be set aside in favor of more effective and less peculiar devices, some of which are already operating at full strength.
It is hard to know what role “moral considerations” actually play in our choices because our behavior is a function of countless known and unknown factors. It is possible that when some of us make decisions, moral considerations are completely idle. But even when moral beliefs do play a role in choices, morality is flexible enough to be available to support any choice anyone is likely to want to make, including the choice of government officials to suppress what they choose to call immorality. This throws a different light on the idea that the moral overlay is useful when we want to justify social and legal sanctions, the second of the reasons Mackie gave for preserving our error-infested moral judgments and language. What good is morality if it can so readily be marshaled to defend the sanctions of a tyrant?
After offering this brief and unconvincing rationale for exploiting the language of morality, Mackie admitted that “a case can be made out for the view that though what I am calling the basic practice is beneficial, it would be better without the moral overlay.” The case he presented is impressive, and Ian Hinckfuss has even claimed that Mackie’s remarks here were an expression of “doubts about the usefulness of the moral institution” that he had begun to entertain “shortly before his untimely death.” In any event, the case Mackie constructed for what we are calling “abolitionism” involves three charges: morality inflames disputes and makes compromise difficult, it preserves unfair arrangements and facilitates the misuse of power, and it makes global war possible.
(1) Morality inflames disputes because moralizing an issue tends to excite and confuse the parties involved. If we hope to resolve conflicts by arriving at a compromise, our task will be easier if moral disagreements are seen as partial conflicts of interest “without the embroidery of rights and moral justification.” The controversy over abortion would not be nearly as intractable as it has become if the fiction of moral rights had not been appropriated by both sides. If the issue is not moralized, Roe v. Wade looks like a sensible compromise between two extreme positions, but when the right to life is set against the right to choose, neither side can yield without violating morality. A human embryo is what it is, but someone who insists on describing it with morally loaded terms like ‘person’ or ‘innocent human baby’ leaves no room for compromise over issues like abortion or embryonic stem cell research. How can anyone compromise with someone they see as wanting to murder babies?
Not only does the moral overlay inflame disputes and make compromise difficult, the lack of an actual truth of the matter opens the game to everyone. Every possible moral value and argument can be met by an equal and opposing value or argument. The moral overlay adds an entire level of controversy to any dispute, and it introduces unanswerable questions that usurp the original question, which is always some practical question about what to do or support. This “moral turn” guarantees that the participants will be distracted from the real issue, and that the disagreement will flounder in rhetoric, confusion, or metaethics. The dangers of the moral overlay are far worse than Mackie thought.
(2) The second problem Mackie found with the moral overlay is that the addition of moral overtones to a practice “will tend to stabilize whatever differential advantages the various parties initially have.” Property laws are made by those with property, and when the laws are given a moral defense, so are the inequalities. Then “the moral overtones of the duty of allegiance, with the associated concepts of loyalty, patriotism, and the like” can be used to condemn reformers as traitors and critics as criminals, and to facilitate the misuse of power. In fairness to the moralist, it should be admitted, as Mackie did, that moral principles can also be called upon to protect us from the abuse of the powerful. But these principles only protect us if those with the power to abuse us accept the idea of morality and are moved by it. It is likely that they do accept morality and that they are moved by it when it tells them what they want to hear; but it is also likely that they will never accept a moral directive as authentic if it threatens to deprive them of any property or privilege they cherish.
(3) Mackie’s third point was that “without morality there might, indeed, be more small-scale fighting; but war as we know it, organized on a national or an international scale, would be impossible.” Mackie did not explain why he thought that without morality we would not have global or international wars. Perhaps he was thinking that we need the extra fictional power of morality to mobilize against a nation of strangers, but not against our neighbors, who always give us concrete causes for annoyance. He had no way of knowing that the threat from terrorists with weapons of mass destruction would eclipse the threat of global war at the start of the 21st Century, nor could he have predicted the damage one or a few individuals might contrive to accomplish in the name of morality. The truth seems to be that morality is invariably called upon to underwrite the actions of both sides of any violent conflict, large, medium, or small, and this does seem to be a reason for thinking that we might be better off without it.
As strong as these considerations for eliminating the moral overlay are, they were not strong enough to turn Mackie into an abolitionist. He admitted that morality has “side-effects some of which benefit some people at the expense of others, while others do more harm than good to almost everyone,” but he avoided a final summation by asking his imagined abolitionist opponent, “What would work in its place?” Answering for that abolitionist, Mackie suggested that the basic practices might be supplemented with the “social psychologist’s techniques of conflict resolution.” Even so, he said, there is a “case for the other side.” The abolitionist’s replacement of the moral overlay with techniques of conflict resolution may not be the best option because it is “easiest to understand the techniques of conflict resolution working within an agreed framework of prima facie—not absolute—rights possessed by all the parties to an issue.”
Mackie may have been inclined to support the institution of morality, but if his final thought on the matter is that that the policy of appealing to moral rights, even prima facie ones, will resolve more conflicts than it nourishes, we may observe that this is not obviously true. We will be better off with prima facie moral rights than with absolute ones, but Mackie still needs to show that it will be easier to resolve practical conflicts when the parties believe in prima facie moral rights than when they believe in no moral rights. Unfortunately, reluctance to compromise can be resolute even when prima facie rights are the only rights mentioned. My prima facie right to peace and quiet is likely to seem stronger to me than your prima facie right to practice your trombone, and the very idea that a right, even a prima facie one, is objective, natural, or God-given, will lead to holding it with that special fondness that makes compromise difficult and violence attractive.
A world where the concept of moral rights has been abandoned is a long way from anarchy. Take away moral rights and there remain legal rights, civil rights, conventional rights, and countless entitlements and liberties people tend to give and demand. It is even possible that we might end up with more happiness or satisfaction if we were to delete the concept of a moral right from our minds. But we will do better still if we can replace that notion with something that works as well or better, and doesn’t require us to believe (or pretend to believe) a host of falsehoods. You don’t need a right to privacy if people have a habit of respecting your need for privacy. It is not hard to think of ways to encourage people to take the needs and interests of others more seriously. We could work harder at teaching and promoting communication skills, which would improve mutual understanding; and we could stop looking at compromise as if it were surrender.
Mackie concluded his case for keeping the moral overlay by saying that morality is less likely to have regrettable side effects if it is understood. But if he was right about what morality is, then when it is understood it will be seen to be the fiction, invention, or projection it is, and those who understand this may no longer be so interested in doing and avoiding things for moral reasons. So it is possible that, contrary to what Mackie said, to the extent that morality is understood it will either fall out of use, or it will be used with even more cynicism and regrettable side effects than it now is.
2. Assertive Moral Fictionalism. In a recent paper, Daniel Nolan, Greg Restall, and Caroline West endorse Mackie’s characterization of our deep moralist tendencies. They say that most people do believe that there is a realm of objective moral facts with prescriptive and motivational force. They also agree that when we make moral judgments we are saying things that are “literally false.” Nevertheless they claim that there are practical advantages to continuing to use moral language and they recommend that we reject abolitionism, which they call “eliminativism,” and characterize as the view that “we ought to abandon false talk of rightness and wrongness, goodness, and badness, duties and obligations, justice and injustice altogether.” Moral fictionalists, they say, believe that “realist moral discourse should be retained, even though it is strictly speaking false, because it is useful.”
Like Mackie, Nolan, Restall, and West offer several considerations to support their recommendation. They say that moral judgments help us coordinate attitudes and regulate interpersonal conflict, but our discussion of Mackie’s reflections suggests that the opposite may be true. If both sides can call upon morality to ratify their positions, then their attitudes will continue to clash, and the conflict is likely to increase. If we hope to resolve conflicts by arriving at a compromise, our task may actually be easier if moral disagreements are seen as conflicts of interest without, as Mackie said, “the embroidery of rights and moral justification.” By embracing moral fictionalism, error theorists regress to the comfortable familiarity of normative ethics, pretending to seek answers where, as they already know, no answers are available. Nolan, Restall, and West say that “the institution of realist morality has taught us to discuss the resolution of such matters in terms of rights, duties, and obligations, and to abandon these frameworks for discussion would deprive us of many of the procedures and tacit understandings that provide a well established framework within [which] such discussions take place.” Exactly; but one of the advantages of eliminativism is that it frees us from that well-established framework and those tacit understandings that lock us into interminable arguments, ultimate impasses, righteousness, rhetoric, and error.
The major argument used by Mackie and by Nolan, Restall, and West, is that we should keep the moral overlay because we are better off with it than we are without it. But to this Nolan, Restall, and West add that morality is such an established part of our way of dealing with the world that it would be “difficult and inconvenient” to do without it. This, of course, is not a very good reason for holding onto a false belief, or for encouraging others to do so. It was difficult to abandon geocentrism and the belief in the efficacy of human sacrifice, but we are better off for having done so. Actually, how difficult or inconvenient it will be to abandon the moral overlay, or any other set of false beliefs, will depend on how rigid and indoctrinated we are, and on whether we can find other ways to say what has to be said and do what has to be done. It might be difficult for some dedicated moralists to break the habit of moralizing, but not everyone is so inflexible. Contrary to what Nolan, Restall, and West suggest, the changes we would have to make if we stopped employing the moral overlay are not nearly as great as those that would be required of us if we decided to abandon the language of folk psychology.
Others (who apparently have not tried it) make a point of how difficult it might be to break the habit of moralizing. William Lycan says that “to produce a genuine freedom from moral intuitions, one would need a steady diet of hard drugs, or some other very powerful alienating force.” Richard Joyce quotes Peter Singer, who thinks that even if we avoid all moral language “we will find it impossible to prevent ourselves inwardly classifying actions as right or wrong.” Joyce observes, and I agree with him, that these remarks are overstated. Nolan, Restall, and West, emphasizing the tenaciousness of the habit of moralizing, mention a kind of “inconsistent nihilism” where error theorists “find themselves unwilling or unable to refrain from making positive moral judgments.” There may be some people who are addicted to morality, but almost any habit can be broken. Even if, as Joyce argues in The Evolution of Morality, there is a genetic basis to think morally, we have managed to alter and suppress behavior like territoriality and aggression that are also grounded in dispositions based on natural selection.
Another benefit of moral fictionalism alleged by Nolan, Restall, and West is that it “enables us to avoid raising complicated metaethical questions every time we wish to discuss an applied ethical question.” But in reality, the opposite seems to be true. Any time we find ourselves involved in some difficult question of applied ethics, questions of metaethics inevitably arise. It is often reasonable to ask if some disagreement may be based on a semantic misunderstanding, and it is customary for both sides of any moral argument to display and demand reasons. So how does fictionalism bypass metaethics? The best way to avoid metaethical questions, or at least the ones we cannot answer, is once and for all to reject normative ethics— that is, to become an eliminativist.
No one is likely to offer moral fictionalism as a descriptive account of what people now mean or think when they make moral judgments. For one thing, most people are not moral error theorists. As Nolan, Restall, and West understate it, the descriptive option “is likely to be implausible.” They are not saying that we are all error-theorists intent on preserving the moral overlay, they are recommending that error theorists continue to speak as moralists, just as the eliminativist or abolitionist is recommending that error theorists stop using the language of morality.
When Mackie defended his version of moral fictionalism he was thinking of our actual world and advising the rare error theorist to continue to operate with, and even improve upon, the moral overlay. Nolan, Restall, and West agree that when we make moral judgments we are making assertions that are “literally false,” but it is not clear what sort of world they are talking about. In arguing for the value of their fictionalism they imagine a case where “all the parties to the debate are moral fictionalists.” Does this mean that we are to try to answer the question of whether moral fictionalism is sufficiently useful by imagining a world where everyone, or nearly everyone, is a moral fictionalist, and then asking if that world generates more satisfactions than one where everyone, or nearly everyone, is a moral abolitionist? It is hard to know even how to begin such a thought-experiment, but we can be sure that of those who try it, some will favor the world of fictionalists and others will disagree.
3. Assertive Moral Abolitionism. The assertive moral fictionalist is an error theorist who urges us (or at least urges other error theorists) to continue to assert moral judgments, and to speak and argue as we would if we were confident moral realists. Even though we may not believe these moral judgments, many of those who hear us will, and, we may hope, this will motivate them to adhere to the fundamental social practices we value. Assertive moral abolitionists are error theorists who recommend giving up moral judgments, which they construe as assertions, because they believe that the harm that flows from asserting and supporting false moral beliefs outweighs the good that results from exploiting them.
If the question of whether we are better off with or without morality is an “empirical” question, it is one to which we are not likely to find a definitive answer. There will be plenty of opinions, but the future is unknowable and our imaginations are both limited and hostage to our desires. That said, it is also true that we can sometimes tell when an idea (revenge) or an institution (slavery) is likely to have bad consequences. After rehearsing their reasons for recommending that we keep the moral overlay, moral fictionalists often acknowledge that morality has, as Mackie conceded, some side-effects that “do more harm than good to almost everyone,” but they do not go into this very deeply (or at all). One error theorist who did accentuate the negative was Ian Hinckfuss, who strongly defended the abolitionist view that we would be better off without morality.
Hinckfuss claimed that we live in a “moral society,” that is, a society in which almost everyone does accept the literal truth of the moral overlay. We are constantly making moral assertions, but they are all false. He acknowledged that false moral beliefs may have some benefits, but argued that any moral society will come with a number of disagreeable features that are essential to its propagation and preservation, features like “elitism, authoritarianism, guilt complexes, ego competition, economic inequality and war.” Moral societies, he added, are intellectually dishonest, “inefficient in maximizing human happiness, satisfaction, or self-esteem,” and “because of the threat of war with other societies, physically dangerous.” The moral overlay actually hinders the resolution of conflicts and fosters the exploitation of the “poor and the weak by the rich and powerful.”
In a moral society children will be raised in “an environment of continual moral injunctions,” and they will be conditioned to want to be good, and trained to respect the moral authorities and their values. When they reach this point they are “in a position to be morally propagandized by those whom they regard as their ‘betters’, that is, those who they feel know more about what is right and what is wrong than they do.” Ordinary members of a moral society will not be able to find the moral truth (because there is none), but they will have learned to believe in moral truth, and to rely on the members of a moral “elite,” who take themselves to be superior in knowledge, virtue, and worth. When one is an honored member of a moral elite “it is easy to believe that what one wishes for oneself is morally permissible, and how one wants others to behave is morally obligatory.”
Hinckfuss is more emphatic than Mackie was about the tendency of those with strong moral beliefs to go to war. After mentioning a dozen or so of the world’s bloodiest conflicts, he asks us to “think of how the situation would have been if, by a miracle, moral thought could have been eradicated from the minds of all the agents involved. . . . There would be no sense of duty, no sense of loyalty, no patriotism, no feeling morally obliged to fight for a cause, no sense that the people one is trying to kill or subjugate are less worthy of survival or freedom than oneself or anyone else.” Joyce replies to this by pointing out that morality didn’t cause these wars, which is right, but he neglects the fact that morality made it more difficult to resolve the conflicts that led to them, and the fact that once a war starts, morality can be used to justify inflicting any cruelty deemed necessary for victory.
Getting rid of morality will not solve all the world’s problems, and no moral abolitionist would claim that it would, but it will allow us to see conflicts of interests for what they are and other people for who they are, and that by itself will undermine demagoguery and fanaticism. If Hinckfuss is even partially right about what it takes to institute and enforce morality, it may be a far more dangerous institution than its fictionalist supporters have allowed.
Although he thought that all of the concepts of morality are dangerous, Hinckfuss found the notion of moral desert particularly harmful. We have been conditioned to think that some people are better than others, and that because of this they deserve the advantages they have acquired. These aristocrats and masters have “a vested interest in preserving the system,” and they have the power to do so. If somehow we could manage to “discard” the idea that some people deserve their unfair advantages, or better, if we could abandon the doctrine of moral desert entirely, “the perpetuation mechanism of morality would be lost and morality itself would rapidly become non-existent.” This, Hinckfuss says, “is the bloodless . . . revolution that I, for one, would welcome.”
4. Practical Rationality. If we do decide to abandon our false moral beliefs and to refrain from using moral categories and language, we seem to be left with “practical rationality.” We must figure out what is in our short, middle, and long-term interest, and base our decisions on that. Practical rationality might tell us that widespread use of the moral overlay is in our self-interest, or it might support abolitionism. No one can say for sure; but no one who is interested in achieving his or her goals is a stranger to practical rationality. Moral abolitionists have no desire to abolish it, and neither do moralists or moral fictionalists. The real question is, “Is practical rationality enough, and if not, what else do we have?”
It is possible, and perhaps even likely, that practical rationality is another fiction. We have never been very good at calculating consequences, and usually the proponents of each side of a dispute predict those consequences that bolster their own positions. The same complex of computational problems that keeps utilitarians from ever being sure of their bottom lines also undermines those who dream about calculating costs and benefits. How, for example, do we weigh the claims of our short, middle, and long-term goals? Nevertheless, if practical rationality is a fiction, it is less obviously a fiction than is morality. Sometimes when we predict a disaster, disaster comes, and some things that appear to be in our interest really are in our interest.
Joyce says that Hume’s answer to the sensible knave is “roughly correct.” It may be in our interest (almost all of the time) to refrain from “defecting,” and even when it appears that we can get away with some breach of trust, we can’t be sure, and the penalty for a mistake is often great. So we usually do have “good instrumental reasons for acting in a cooperative manner.” But, he adds, this answer would only “be good if we lived in a world populated entirely by humans enjoying perfect rationality.” Since we don’t, practical rationality, a concern for our own welfare, will not prevent all defections. So, Joyce concludes, we need to supplement practical rationality with the fiction of morality, but not by embracing or promoting, or even asserting, false moral beliefs. His moral fictionalist will still be an error theorist, but one who expresses moral judgments without asserting them and who thinks moral thoughts but has no moral beliefs.
5. Non-assertive Moral Fictionalism. If we add up the negative aspects of the moral overlay mentioned by Mackie and Hinckfuss, and then factor in the criticisms of the arguments brought forward in its support, we do have a case against using it. But in addition to those considerations, we also must include the easily predictable disaster that will result from encouraging everyone to assert and defend propositions known to be false. Even if the institution of morality offers many practical benefits, a policy of promoting false moral beliefs will undermine clarity, destroy trust, increase cynicism, and reward sophistry. If we want the benefits of the moral overlay without the damage spawned by the ill-advised worship of false beliefs, we need to find some way to make moral judgments without making or believing false assertions. This is the aim of “non-assertive fictionalism,” a development of the error theory explored by Richard Joyce, who sees this position as one last way an error theorist might be able to reject the moral abolitionist’s disconcerting recommendation.
The non-assertive moral fictionalist, says Joyce, wants to continue with the moral overlay because he believes that “morality, though in error, is a justified practice in light of its usefulness.” Joyce writes as if when the truth of the error theory dawns on us it will be relatively easy to continue speaking and even thinking as moralists. He speaks of someone who has become an error theorist but who is so used to employing the concepts of moral right and moral wrong that she carries on using them in everyday life and “is happy to do so.” But getting clear about metaethics is hard work, and an actual insight into the truth of the error theory is revolutionary and not easily put aside. A clear-headed and convinced error theorist is not likely to fall mindlessly into old habits of moral speech, so continuing with the moral overlay will be a decision, at least at first, to make and defend what one believes to be false assertions about value and obligation. If Fred, newly converted to the error theory, decides to continue using moral language with his moralist friends, he will not be slipping into a kind of non-assertive fictionalism because he will know that his moralist friends will take him to be making full strength moral assertions. This means that he can not even intend not to be making assertions.
Joyce says that Fred, moralizing in the midst of moralists, may fail to assert because “the conventional understanding between speaker and audience has broken down to such a degree that Fred’s speech acts are . . . ‘misfiring’.” Fred, he adds, has “ceased to speak.” But unless Fred has completely lost touch with the existing conventions, and with the fact that willingness to participate in those conventions is signaled by the straightforward utterance of a non-bizarre descriptive sentence, he knows that he will be understood as asserting. So he is asserting.
But now suppose that Fred does somehow forget or repress his understanding of the error theory. Imagine that, as Joyce says, Fred has pushed his former acceptance of the error theory so far to the back of his mind that moral thoughts like “But it would be wrong” seem like beliefs to him. He says that Fred’s “well-rooted” moralistic habits are likely to be brought up short only when he “enters a very critical context of discussion, such as a conversation on metaethics.” Joyce may believe that these metaethical excursions would be rare, but I suspect that questions about meaning and justification will turn up as soon as we disagree about anything that really matters to us. Since very many things do matter to us, our frequent metaethical digressions will remind us of our belief in the error theory, and that will make it increasingly difficult to stay immersed in our fiction.
If these metaethical episodes do not alert Fred to the objective (as opposed to the fictional) situation, if he continues to ignore his former metaethical insights, if he begins to have moral feelings, moral outrage, moral guilt, and moral arguments, then we have every reason to say that he has reverted to his former moral beliefs, and to the error he once identified and abandoned.
So how can Fred become a non-assertive moral fictionalist? How can he say things he knows to be false and yet avoid making false assertions? One way to utter a sentence without asserting it is to give some sign that neutralizes the assertive force of one’s utterance. A sarcastic tone, a wink, or extreme exaggeration will withdraw assertive force and set an interpretation problem for the listener. But unless he relies on some such device, Fred cannot unilaterally withdraw assertive force from his utterances.
The other way Fred can manage not to assert the sentences he utters, says Joyce, is for him to become a noncognitivist about his own moral discourse. Noncognitivism “might become true if we were to alter our attitude towards moral discourse.” Then we would be able to utter a sentence like “That was morally wrong,” not as an assertion, an expression of a belief, but “as an expression of a thought.” The content would remain the same, namely the false claim that an objective wrongness qualifies the act, but we would be uttering the sentence with a non-assertive force (or with no force).
This can be done, but not unilaterally. Fred can address moral utterances to others without asserting them if he is participating in a convention that make this possible. As Joyce notes, “whether one asserts . . . depends upon a framework of linguistic conventions within which one’s utterance occurs.” The non-assertive moral fictionalist looks forward to the day when we are all (or almost all) error theorists, devoted to the moral overlay and committed by convention to understand moral utterances not as assertions but as expressions of thoughts, or emotions, or whatever. This might come to pass, but it is not easy to see how we could get there from our present deeply moralistic society, or to see very clearly what “there” might look like. Perhaps we can begin to make sense of a world filled with non-assertive moral fictionalists by considering a group of error theorists for whom the idea of moral fictionalism has become an option. Only in such a group could a convention develop that would allow speakers to unhook their moral utterances from assertive discourse.
6. A Group of Moral Fictionalists. Becoming a non-assertive moral fictionalist is a three step process. The first step is to accept the error theory, the second is to choose to continue with the moral overlay, and the third is to form, or to become part of, a group with a convention that enables one to say things without asserting them. A group of refugees from a seminar in metaethics who have decided to embrace the error theory could agree to speak in moral terms with each other, and agree that in doing so they were expressing but not asserting thoughts. But if these same error theorists express moral judgments in books, or in public, or to “outsiders,” those moral judgments will correctly be taken to be assertions of belief. If we are to be non-assertive moral fictionalists we will either have to speak morality only with the members of our group, or bring it about that the larger group accepts a convention to the effect that uttering a moral judgment triggers a withdrawal of assertoric force.
When Fred, our newly minted error theorist, looks around, he will find that almost everyone (including the moral realists and both the assertive and the non-assertive moral fictionalists) will be deeply involved in the moral overlay, and will be uttering moral judgments, engaging in moral arguments, expressing moral outrage, and making moral demands. If he is surrounded by non-assertive moral fictionalists among whom expressing a moral judgment is understood not to be making an assertion, how will he ever find this out? It will not be apparent from anything anyone says or does. If he asks someone who happens to be “immersed” in the fiction, that person may fail to admit to their disbelief and deal with the question as a genuine moral realist would. If indeed it is possible to be so out of touch with what one believes, which I doubt, then that forgetful error-theorist might even say “Of course I really believe it is morally wrong to torture innocents for fun.”
7. Problems for Non-assertive Fictionalism. Belief vs. Thoughts and Assertion vs. Expression. Non-assertive moral fictionalists are error theorists who set out to express moral judgments without asserting or believing them. In this way they hope to reap the advantages of belief in morality and to avoid the bad effects of asserting and having to defend false beliefs. Joyce says that we can use the moral overlay in our day-to-day life as much as we please, but as long as we would deny it in our most critical moments, we do not assert the content of our utterance, nor do we believe it—we just think it. The difference between a thought and a belief, he says, is the disposition to dissent. As long as Fred once embraced the error theory (in a critical mode—perhaps in a philosophy class), and as long as he might, under similar or different circumstances, do so again, we don’t have to admit that he did have moral beliefs in the interim.
That is an unusual interpretation of belief. The distinction between thoughts and beliefs is real enough, but pushing it to allow the nonassertive moral fictionalist to say that Fred didn’t have moral beliefs even while he was thinking and acting as if he did seems unnecessary. It is the thinking and the acting that cause the trouble and that lead to the endless arguments and intensified conflict. The bad effects of believing in moral facts are as likely to result from the pretence as from the belief, especially since it is sometimes difficult to distinguish the one from the other. Perhaps we can say this: the deeper we go into our moralistic trances, the more likely we are to experience the full range of the bad effects attributed to moral belief; and the less we immerse ourselves in the moral fiction, the less likely we are to see any of its alleged beneficial effects.
What now of those error theorists who continue with the moral overlay but do not lose themselves and revert to moral belief? This group will include assertive moral fictionalists like Mackie, but also non-assertive moral fictionalists who do not intend for their moral utterances to be (or to be taken as) expressions of moral beliefs. We know that if the conventions for doing so are in place we can use moral language without making moral assertions. But why would a group of self-conscious and mutually aware error theorists revisit their moralist past? Assertive moral fictionalists at least believe that others are open to moral persuasion, but if you know that your audience doesn’t believe in an objective morality any more than you do, what is the point? Falling back into the moral framework requires an immense expenditure of time and effort on a project both you and your target audience know to be doomed. There may be differences between moral beliefs and moral thoughts, but when they go public they look alike and they both need to be defended. If we defend them with vigor, then we will be indistinguishable from the moralists, and that will facilitate the return of those interminable debates and ritualistic arguments. But if we defend our fictive moral utterances and thoughts half-heartedly, we will show others and remind ourselves that we are not really serious about what we are saying.
Joyce claims that non-assertive moral fictionalism is useful in overcoming lapses due to weakness of will. He observes that when we are tempted to act in ways that are not in our interest, moral beliefs might help us do what we know we (practically) ought, but, he adds, so can moral thoughts. Just thinking of a fire destroying one’s house and possessions might cause anxiety or fear. Similarly, just repeating or hearing the familiar language of moral obligation might trigger a familiar impulse to obey, and that may be enough to get us to resist some temptation or to eat our vegetables. Apparently the magic works on others in the same way. If I tell you that something is morally wrong, that very thought might elicit in you a mild but negative emotion against the act, even if we are both nonassertive fictionalists sharing an assertion-canceling convention.
A “mild” emotional reaction to a moral thought may add some subliminal motivation, but it is not likely that it will be strong enough to combat an impulse that has already managed to defy our self interest. Joyce says that even if the difference were minimal, “maintaining the moral discourse . . . would at least be a reliable improvement on straight thinking (and straight talk) about one’s long-term preference satisfaction.” This may be so, but moral discourse is not an improvement on straight thinking and talk that ranges a little more widely than a single person’s long-term preference satisfaction. Many of us automatically factor in the interests of others we know and cherish, and sometimes we give weight to the interests of complete strangers. A vast number of known and unknown factors resolve into our decisions: thoughts, beliefs, fears, desires, emotions, habits, memories, resolutions, impulses generated by chemicals and hormones or by what someone has just said to us. In this rushing causal stream, it is unlikely that the echo of emotion from a moral thought will find a way to play much of a role in our actual decisions.
8. A Personal Choice. We may never be able to find an indisputable answer to the question of whether the moral overlay helps us more than it harms us, but it should be clear that there are more problems with the moral overlay than moralists usually admit, and so the question remains: “What is an error theorist to do?” We may or may not have an opinion about whether we would all be better off if most people were abolitionists, but the question that an error theorist cannot avoid is whether to accept the recommendation of the moral fictionalist or that of the moral abolitionist. An isolated error theorist cannot decide unilaterally to stop using moral language assertively, but he or she can decide unilaterally to stop using moral language. Perhaps error theorists will someday blaze a trail to nonassertive moral fictionalism by spreading a convention that cancels the assertive force of a moral judgment. But that day is not likely to come in our lifetimes, so the practical question for an error theorist remains whether to abandon the day-to-day use of the moral overlay. This is a question we can understand, and, unlike the question of whether “we” would be better off if (almost) everyone were an abolitionist, there is a simple and straightforward way to deal with it. We can just cut back on the use of the moral overlay and see how things go.
Even if we are error theorists who aren’t ready to give up the moral overlay, it is easy enough to try out abolitionism and to see what it feels like to minimize our moralizing. We just have to learn to watch ourselves, and to anticipate and then restrain the impulse to moralize. Doing this is no more difficult than teaching ourselves not to swear, and it is not nearly as difficult as getting rid of an accent.
If we are error theorists, and if we can refrain from falling into moralism or any kind of moral fictionalism, then it will soon become clear that we are in no way limited in our ability to communicate our attitudes, requirements, and feelings. Instead of telling others about their moral obligations, we can tell them what we want them to do, and then we can explain why. We can express annoyance, anger, and enthusiasm, each of which has an effect on what people do, and none of which requires duplicity or moralizing.
Non-assertive moral fictionalists, by not promoting false beliefs and assertions, avoid some of the difficulties that confront assertive moral fictionalists. But if too many moral error theorists speak as if they were moralists, then there is a puzzle about how a group of nonassertive moral fictionalists could even form, let alone expand to the point where a significant majority of speakers shared the relevant assertion-canceling convention. Since the majority of users of moral language are currently moral realists, non-assertive moral fictionalism is not at this time a live option. Those of us who have arrived at a belief in the moral error theory do have a choice, but it is between assertive moral fictionalism and moral abolitionism.
We have already been exposed to the case against assertive moral fictionalism, and if we conduct the experiment I suggested, we will soon see that it is neither difficult nor dangerous to give moral language a rest. The moral abolitionist is equipped, as we all are, with a variety of habits, preferences, policies, aims, and impulses that can step up to take the place usually thought to be occupied by moral beliefs and thoughts.
Moral beliefs may have some effect on what we and others do, but less than we think because of their malleability, and because of our skill at rejecting or ignoring ones that we find inconvenient. Moral thoughts are likely to be even less useful. Personal habits, preferences, policies, and principles are not backed by the fictional authority that moral judgments are believed to possess, but they are quite functional and they have the advantage of not being based on error and confusion. By the time the truth of the error theory fully dawns upon us, completing the transition to these purely non-moral considerations is likely to be a simple matter. As we do this, we will find that there will be less to argue about, and that our conflicts and disagreements with others, at last seen for what they are, can be addressed and resolved. The moral overlay may have been useful in the past, and it may or may not be good for society in the future, but if someone who sees it for what it is can discard it, that person can be a witness to the fact that life without an ever-changing and controversial answer book is preferable to pretending to believe what we know is false, and healthier than allowing ourselves to forget our hard-won insights about what is really going on when people make moral judgments.