Moral non-cognitivism denies that moral judgments express truth-apt propositions. Instead, 'stealing is wrong' expresses a negative attitude, endorsement of a norm, or prescription—not a claim about the world that can be true or false. This explains why morality seems meaningful and binding without requiring objective moral facts.
Consider how moral language functions: sometimes it guides action ('don't hurt people'), sometimes it expresses disapproval ('what you did was despicable'). Non-cognitivism argues these functions don't require truth-conditions the way scientific statements do.
From your study of moral language and meaning, you know that moral sentences like "stealing is wrong" look grammatically similar to factual sentences like "the door is open." Both have subject-predicate form; both seem to describe something. But non-cognitivism argues this surface similarity is misleading. The key insight is that language can do many things other than state facts: it can express emotions, issue prescriptions, endorse norms, or commit speakers to courses of action. Non-cognitivism holds that moral sentences perform these other functions and therefore do not express truth-apt propositions—claims that are true or false in the way "the door is open" is true or false.
The earliest non-cognitivist view, emotivism (associated with A.J. Ayer and Charles Stevenson), holds that moral sentences simply express emotional attitudes. "Stealing is wrong" means roughly "boo to stealing!"—it vents a negative feeling and invites others to share it. This explains why moral disagreement is so intense and persistent: it is not a disagreement about facts that could be settled by evidence, but a clash of attitudes. R.M. Hare's prescriptivism developed a more sophisticated version: moral sentences are universal prescriptions. "You ought not steal" prescribes a course of action for all relevantly similar agents, including the speaker. On this view, moral sentences do have logical structure—they can be universalizable, consistent, or contradictory—even though they lack truth conditions.
The most powerful challenge to non-cognitivism is the Frege-Geach problem: moral sentences seem to retain their meaning when embedded in complex logical contexts where they are not being asserted. Consider "If stealing is wrong, then getting your little brother to steal is wrong." The word "wrong" appears in the antecedent, but you are not there asserting that stealing is wrong—you are using it in a conditional. If "stealing is wrong" merely expressed a boo-attitude, it is unclear what it would mean to put it in a conditional. This problem drove later non-cognitivists toward more sophisticated accounts. Quasi-realism, developed by Simon Blackburn, attempts to show that non-cognitivists can progressively earn the right to talk about moral truths, moral facts, and moral knowledge—not by abandoning the expressivist starting point, but by explaining how these ways of talking are legitimate expressions of a practice of endorsing and reasoning with norms.
The non-cognitivist position occupies a distinctive space in metaethics. It avoids the metaphysical difficulty of postulating mysterious moral facts (as realists do) while preserving moral language's action-guiding force. It is not nihilism: non-cognitivists are not saying morality is meaningless or that anything goes. Rather, moral discourse has its own standards of coherence, its own forms of disagreement, and its own capacity to be done well or badly—all without those standards needing to be read off from a moral reality "out there." The challenge the view must meet is explaining how all the logical and epistemological features of moral discourse can be reconstructed within an expressivist framework.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.