3.1.10: Cognitivist and Anti-Realist Theory One (Moral Error Theory)

Thus far, we have seen that Cognitivism tends to be associated with Realism. Mackie breaks with this trend with his Moral Error Theory. Mackie accepts that our moral utterances are expressions of truth-apt beliefs, but denies Realism. In so doing, Mackie denies that possibility that our truth-apt beliefs are ever true, because a moral description of the world can never accurately describe a world without any moral properties in it.

In Mackie’s own words, “Although most people in making moral judgments implicitly claim, among other things, to be pointing to something objectively prescriptive, these claims are all false”. By prescriptive, Mackie means action-guiding and Mackie denies that any objective guides to action (moral properties, in our terms) actually exist.

Mackie’s view is startling and raises loads of questions about how we should live if morality is entirely false. Although interesting, these discussions are not for this chapter. Instead, we must explain and evaluate Mackie’s theory as it stands rather than consider its implications if true. A theory having depressing or liberating implications does not make that theory any more or less likely to be accurate (though it is surprising how often even the best philosophers are prone to such mistaken thinking).

Mackie’s Anti-Realism is supported by the following two arguments. It should be made clear that Mackie’s arguments are directed against both Naturalistic and Non-Naturalistic Realism.

Argument from Relativity

Mackie’s first objection to Realism is built out of his appreciation of the depth of moral disagreement, and so shares something with one of the objections to Intuitionism offered in the previous section. Mackie suggests that in other plausible realist disciplines, such as the sciences or history views begin to coalesce around the truth over time and disagreement is, at least in part, conquered.

Disagreement occurs in these disciplines because there is a barrier to true knowledge and scientists and historians will sometimes, through no fault of their own, be blind to the facts. However, sometimes the facts become clear and disagreement thereby reduces.

Yet, in ethics, philosophers still disagree over the same issues that they were arguing over 2000+ years ago, questions such as “when is war acceptable” and “when can promises be broken”. If moral truths really did exist and Realism was correct, should we not have expected to find some of these truths by now? Thus, Mackie views disagreement in ethics — deep disagreement that seems impervious to solution through rational means — as evidence that Realism is incorrect; there are no moral facts to settle the debates or at least some of those debates would have been settled by now! Of course, if you think that some moral debates have been settled, then you could use this to criticise this Mackian argument.

Argument Queerness

Mackie’s second anti-realist argument is his most famous. Moral properties — be they natural or non-natural — are supposed to be action-guiding. If it is true that murder is wrong, then we should not murder, even if we might want to. Equally, if it is true that giving to charity is right, then we should give to charity, even if we might not want to At its core, morality is supposed to offer reasons for action that we cannot simply ignore even if we like murdering or hate charitable giving. This aspect of morality, however, raises issues at the metaethical level.

David Hume (1711–1776) recognised the potential problem with the action-guiding quality of morality when he spoke of the “is-ought” gap. According to Hume:

In every system of morality, which I have hitherto met, I have always [remarked], that the author proceeds for some time in the ordinary ways of reasoning, and establishes the being of a God, or makes observations concerning human affairs; when all of a sudden I am [surprised] to find, that instead of the usual copulations of propositions, is, and is not, I meet with no proposition that is not connected with an ought, or an ought not. This change is imperceptible, but is however, of the last consequence.

Hume wonders why and how we move from statements about what is the case, to statements about how we ought to act. We do not make such a link between “is” and “ought” in areas other than morality — the fact that a horse is running at Goodwood does not, of itself, give you an “ought” regarding how to act in response. The fact that a moral property is, on the other hand, does seem to give rise to such an “ought” regarding behaviour. How can this be explained?

Hume has his own suggestion for explanation, and this is outlined in section twelve. Mackie, however, takes this Humean worry in his own direction. Mackie suggests that properties themselves that carry such an action-guiding quality, that offer an “ought” just because they are, would be extremely queer properties. He says that “[if] there were objective values [moral properties], then they would be entities or qualities or relations of a very strange sort, utterly different from anything else in the universe”.

Mackie suggests that if we can explain moral thinking without resorting to positing the existence of such queer and utterly unique entities then we would be better off. The simpler explanation is not to grant existence to weird properties, but just to suggest that there are no properties and that our moral beliefs reflect cultural and personal beliefs. Just as we do not tend to suggest that aliens or ghosts exist on the basis of first-hand testimony (competing explanations based on drunkenness or tiredness, for example, seem more plausible) so we perhaps ought not to grant that moral properties exist just because we happen to talk about them.

Indeed, support for Anti-Realism through a complaint about the queerness of moral properties is further supported via consideration of Hume’s fork.

Hume divided knowledge into two camps — knowledge gained from relations of ideas and knowledge gained from matters of fact. Knowledge claims like “2+2=4”, or various geometric claims like “triangles have three sides”, are established in the former way whereas knowledge claims like “Alastair is wearing a blue shirt today” are established in the latter way.

This split of types of knowledge is referred to as Hume’s fork, yet claims to moral knowledge do not seem to fit either side of the fork. Moral knowledge is not derivable simply from relations of ideas (it is not supposed to be like geometric or mathematical truth and cannot be deduced a priori without any testing the world through our senses).

Nor, however, is it derivable simply from matters of fact, given the “is-ought” gap referred to above (a posteriori, sense-based, worldly and scientific empirical observations reveal what is, not what ought to be). If moral knowledge does not fit into either side of Hume’s fork, then it will be the case that either moral knowledge is a completely unique type of knowledge accessed in a completely unique way or, more plausibly perhaps, moral knowledge does not actually exist. But if we cannot know that moral properties exist then we should not be realists.

Hume, certainly, would have rejected the idea that moral properties existed based on the application of his famous fork. Remember, however, that Hume favoured Non-Cognitivism and Anti-Realism rather than (like Mackie) Cognitivism and Anti-Realism.

On a similar theme, Mackie strengthens the argument from queerness by referring to the queer method of understanding that we would need in order to come into contact with queer moral properties. Mackie suggests that we would need a special moral faculty in order to access queer moral properties. Although Mackie admires the honesty of the intuitionist in admitting the existence of such a queer moral sense, he does not think that it is credible to believe in the existence of such a radically different faculty for accessing realist moral properties in the world.

As before, if we can explain our moral beliefs without needing to admit the existence of queer properties, then why admit to the existence of a queer method for grasping queer properties? Moral Realism, according to Mackie, thus requires an unnecessarily queer metaphysics (what exists) and an unnecessarily queer epistemology (how we know what exists). For these reasons, Mackie is an anti-realist.