February 15, 2021

Making Sense of Morality: Error Theory

February 15, 2021/ R. Scott Smith

Editor’s note: R. Scott Smith has graciously allowed us to republish his series, “Making Sense of Morality.” You can find the original post here.

Naturalism and Error Theory

Still another kind of naturalistic moral cognitivism is error theory, which has been defended by J. L. Mackie (d. 1981) and Richard Joyce (b. 1966). Mackie used two arguments for his view. First, the argument from relativity maintains that empirically, there are many moral differences amongst people. For him, the best explanation is that moral judgments are tied to different forms (or ways) of life, each of which has its way of interpreting reality. Second, the argument from queerness says that, if objectively real, moral qualities would be very bizarre things in an otherwise naturalistic world. Since we know natural kinds of things empirically, to know objectively real, intrinsically moral properties would require some extraordinary faculty.

Moreover, moral discourse is dependent upon institutional facts. Unlike brute facts about the world itself, which are natural, institutional facts are social constructs, due to how people in various societies (or forms of life) have constructed their institutions and their rules that guide peoples’ actions, including moral discourse. When people speak as though morals are objectively real and not dependent on their social, institutional settings, they show that moral discourse really is filled with error.

Joyce follows Mackie in that there are no independently real moral facts. Still, as a kind of moral cognitivism, error theorists do maintain that we do make moral judgments. However, since there are no real moral qualities, these claims are systematically false and thus filled with error.

Still, Joyce thinks that error theory does not demand that people give up engaging in moral speech. It is just that, to be consistent, their speech act is only making an assertion. They are not believing these moral claims to be true. Additionally, since error theorists reject the existence of any real moral properties, they deny that any action is moral or immoral. Nevertheless, they still can oppose others’ actions, for that need not require that they believe objective moral properties exist.

Assessment

Error theorists consistently hold that on naturalism, there are no intrinsically moral properties. This naturalistic view of what morals are trades upon language use. They are just ways of speaking according to the “grammar” (or, rules) of a given people that allows them to use moral discourse, yet while (apparently) avoiding the reality of morals.

Now, we will see when we explore ethical relativism that while there is a fact of moral diversity amongst people and cultures, nonetheless those differences may not be as wide or deep as we have been taught. Instead, we can identify common morals that may be applied differently (e.g., how people in one culture show respect for their elders, versus how people in another culture do so). Further, just because there is a descriptive fact of diversity, that alone does not give us ethical relativism, which is a normative thesis.

Granted, too, irreducibly moral properties would be rather “queer” given naturalism. But, perhaps there are independent reasons why we should question that assumption. In later essays, I will suggest a few such reasons.

Moreover, it is true that we may speak in ways that do not necessarily commit us to the reality of things we are talking about. Generally, mere word uses do not have power to cause things to come into existence (except, for instance, stories). A scientific example was talk of phlogiston to explain combustion. Later, however, scientists discovered it was not real; instead, oxygen was what was involved.

Further, error theory does not explain why we find morality to be such a ubiquitous aspect of life. After all, why talk morally if there are no morals? While error theory explains why we can talk morally, given naturalism, it still does not give us an adequate explanation of what morals are. If they are just the way we use words, then we can change morals by changing how we talk. In that case, murder could become right, and justice could become bad. But surely that is false.

For Further Reading

Richard Joyce, “Mackie’s arguments for the moral error theory,” Stanford Encyclopedia of Philosophy

J. L. Mackie, Ethics: Inventing Right and Wrong

R. Scott Smith, In Search of Moral Knowledge, ch. 5

John Hare’s God’s Command, 8.3.3, “Sharon Street”

February 16, 2018/ David Baggett

In 2006 Sharon Street published an article, “A Darwinian Dilemma for Realist Theories of Value,” which has been the subject of a considerable literature in reply. Her argument relies on the primary claim that our normative dispositions—that is, our dispositions to form certain normative beliefs rather than others—are (largely) selected because they have some natural property. For example, perhaps they contribute to reproductive success by promoting certain kinds of cooperation. But from the perspective of realism, accepting this claim defeats our epistemic entitlement to our normative beliefs, because we will come to be aware of the unlikely reliability of the processes that shaped those beliefs.

This is the Darwinian dilemma: the realist has either to deny the primary claim or to concede that her “normative judgments are, by her own lights, irrational.” She’s not arguing for skepticism or for the impossibility of ethical knowledge. Rather, she is trying to show that, if there is to be ethical knowledge, it has to be understood on an anti-realist model. Her point is that all that natural selection needs is our beliefs in the normative facts, not the normative facts themselves. If our normative and theological beliefs are largely the product of our evolutionary history, fitness-enhancing beliefs about morality and gods will be adopted, regardless of whether they are, in the realist sense, true or false. Even if a particular belief is false, it may promote genetic propagation.

This is the challenge. But there is a good response to it. Even if we grant that natural selection has given us normative belief-forming dispositions that are not truth-tracking, and that have in fact given us a mixture of “nasty” belief-forming dispositions and corresponding behaviors alongside other “nicer” ones, and even if we grant that therefore our normative beliefs are unreliable to the extent that they are given to us by natural selection, nothing follows about how many of our normative beliefs are formed in this way.

Consider the analogy with mathematical beliefs. To what extent do we have the ability to track truths about non-linear algebra? The point is that, even if we get our cognitive equipment from evolution, we can use that equipment to reach beliefs that are independent of adaptive value. It remains possible that cultural evolution has been operating to refine our normative stance in a truth-tracking way. If we use the phrase “cultural evolution” loosely, we can make the point that admitting a significant initial effect of biological evolution on belief formation does not license the conclusion that natural selection is the sole force in all our belief formation thereafter.

The initial effect of natural selection is still relevant, because, if we were given cognitive equipment that was hopelessly and permanently vitiated, then we could not hope to use this equipment to discriminate subsequently between the beliefs in the initial mixture that we should endorse and the ones we should reject. We would be, so to speak, fatally handicapped. But there is no reason to think our situation is hopeless in this way.

Are our current normative disposition all simply products of natural selection and not (partly or wholly) products of experience, reflection, and reasoning guided by moral reality as such? This is a metaphysical question, not one proper to science in its own domain. Ruse’s recognition of this separates him from Mackie. We need to distinguish the claims of science and the claims of “scientism,” which is the attempt, as Ruse puts it, to make science say everything. Metaphysical naturalism claims baldly that there is nothing beyond physical reality, but this is a claim that requires philosophical justification and is not within the proper sphere of science. Street’s argument does not give us any reason to believe that metaphysical naturalism is true.

Image: Australopithecus Afarensis, Lucy. C. Lorenzo. CC License.

December 28, 2015

Chapter 5, Part I, C. Stephen Evans’ God and Moral Obligations, “Alternatives to Divine Command Theory”:

December 28, 2015/ David Baggett

God and Moral Obligation by C. Stephen Evans

In this chapter Evans looks at metaethical views that some will see as a rival to a divine command theory (DCT) to see what strengths and weaknesses they have. Some aren’t really competitors, and for those that are Evans will try to show that they face serious objections that a DCT does not face. He will try to select examples of each view that are prominent and representative, without claiming that such views exhaust the territory.

ERROR THEORY

J. L. Mackie was well known for his moral skepticism and “error theory” in ethics. Ordinary morality, he thought, is best thought of as a kind of “folk theory” that turns out to be false. Mackie presents a number of arguments for this view. First, he thought a subjective account of morality accounts for the relativity and variability in moral beliefs and practices. Second, objective moral value would be “queer” in the sense of being peculiar; they have no foundation in the world as described by science. Third, it’s hard to see why moral values should supervene as they do on natural features of the world. Fourth, it’s hard to see how such objective values could be known even if they are real. Finally, a reductive explanation of beliefs about values undermines any claim to objectivity.

How should a DCT’ist respond? Well, she can join her voice with various other ethicists (Kantians, natural law theorists, utilitarians, and the like) to argue for the objectivity of ethics. Beyond that, though, she can show that several of Mackie’s arguments work well against naturalistic theories. Values and other moral properties are indeed queer in a naturalistic world, but not a theistic one. Likewise it would be strange in a naturalistic world that humans have cognitive capacities that give them understanding of the good and the bad, of right and wrong, but not in a theistic one. Interestingly, Mackie himself imagined how God could play a role in ethics much as Evans envisions. Mackie didn’t subscribe to the view, but he thought it coherent and could see how it could defuse the Euthyphro objection.

Nietzsche, another atheist, similarly saw ethics as connected with God. His scathing critique of secular ethics was based on the way it tended to assume objective morality is possible without God, which he thought ludicrous. In this way he offered the testimony of an “unfriendly witness” that objective moral obligations require God and make sense only, or at least the most sense, if God exists.

EXPRESSIVISM

Expressivism as a metaethical theory comes in a variety of forms, from the emotivism of Ayer to the sophisticated quasi-realism of Blackburn. What they hold in common is “non-cognitivism” or “anti-realism”: the rejection of the idea that moral propositions express objective truths. Instead moral statements express emotions (Ayer), attitudes (Stevenson), prescriptions as to how one should behave (Hare), plans to which one is committed (Gibbard), or perhaps a complex mix of such subjective states (Blackburn).

The strength of the expressivist view is that it appears to account for why morality matters, and why moral claims can motivate as they do. It links to our actions. But Evans wants to raise a question about whether it links morality to behavior in the right way. The question he wants to raise is not whether moral judgments can motivate, but whether on expressivist views such judgments can have the kind of authority morality ought to have.

Many early criticisms of the view were based on the claim that such views do not seem to do justice to moral disagreements and arguments. Relatedly, Geach said it couldn’t make sense of moral propositions figuring in logically valid arguments. This led to more sophisticated accounts. At the heart of such views lies the idea that even though moral statements do not express propositions with genuinely objective truth values, there is a natural human tendency to “project” our emotions, attitudes, prescriptions, plans, etc. onto the objective world. This projective theory gives a reductive explanation of why moral language has the features it does that enable moral statements to mimic propositions that have genuine representational content. Blackburn and others have in turn developed accounts of the “logic” of moral statements that explain how it can be that these statements mimic the properties of genuinely representational propositions, even though they actually don’t refer to anything.

Evans thinks the real difficulty with the view lies with the way that expressivism, even in its projectivist, quasi-realist form, undermines the authority of moral judgments, especially judgments about moral obligations. Take emotivism, for example. Why should Mary care about the approval of James? One might think the problem is that the James doesn’t mean enough to Mary, but that’s not really the point. The challenge is to account for moral authority. The more sophisticated quasi-realism of Blackburn may appear to help with this problem, but the help is illusory. For in the end moral judgments merely mimic statements that can be true or false independently of the stance of the person making the judgment.

Blackburn doesn’t think his view makes truth relative, because if we “step back into the boat,” as it were, and put back the lens of a sensibility, there’s nothing relativistic left to say. Evans replies, though, that for the person who has awakened to the truth of projectivism, even this will be difficult to do or even impossible for some. How can we get back into one particular boat and believe that it’s the “right” boat, when we know there’s no such thing as the right boat?

If we could segregate our beliefs about normative ethics from our metaethical beliefs, perhaps Blackburn’s view would work, but it is not easy to wall off our beliefs about morality from our actual moral convictions. In the end, quasi-realism is a form of moral skepticism, only Mackie’s theory is transparent and honest, while the skepticism on Blackburn’s part is disguised by the fact that he continues voicing some elements of his own moral stance as if they were objectively true judgments. But the truth on offer seems a pseudo-truth, a “semantic shadow” of the attitudes and stances taken by ordinary people.