Moral Cognitivism and Noncognitivism

The Philosophical Problem of Moral Judgment

A moral judgment is an evaluation of conduct, character, institutions, obligations, or states of affairs under concepts such as right, wrong, good, unjust, admirable, and obligatory. The central metaethical question is what a speaker does, thinks, and commits herself to when she judges that a deed is wrong. An assertion of wrongdoing may resemble a factual report: it can be challenged, offered as a reason, embedded in a conditional, and conjoined with other assertions. Yet moral judgment also appears intimately related to approval, disapproval, choice, condemnation, and demands upon action. The combination generates the enduring controversy about moral cognitivism and noncognitivism.

A cognitivist takes ordinary moral judgments to express beliefs with propositional content capable of truth or falsity. The specific properties allegedly represented, the conditions of truth, and whether any such judgment is actually true remain open questions among cognitivists. A noncognitivist emphasizes the practical attitudes expressed through moral judgment and, in its characteristic formulations, challenges the idea that the primary semantic function of moral discourse is to represent moral facts. Contemporary versions often retain deflationary truth talk while proposing a fundamentally attitude-based explanation of moral content (van Roojen, 2023, SEP).

Consider the sentence "Deliberately humiliating a vulnerable person is wrong." A cognitivist analyzes its acceptance principally as belief that the described conduct has a morally relevant property. A simple emotivist analyzes its utterance principally as the expression of disapproval. A norm expressivist analyzes acceptance by appeal to a stance about what conduct ought to be ruled out. Each theory must explain what happens when the sentence appears under negation, within a hypothetical, as the conclusion of reasoning, or in a disagreement whose parties take one another to be mistaken. Its explanatory task extends beyond the isolated affirmative example.

The dispute is important because theories of moral knowledge, practical reason, moral objectivity, and public accountability routinely presuppose an answer about the nature of judgment. The study accordingly treats cognitivism and noncognitivism as competing explanatory programs rather than as mere labels for ethical optimism and skepticism. Its distinctive object is the relation among moral language, mental state, and logical and practical commitment; the broader arguments for moral realism and the substance of competing first-order ethical codes belong to their own research questions.

The Dimensions of Cognitivism and Noncognitivism

The debate can be mapped through five distinguishable questions. The semantic question asks whether a moral sentence expresses a proposition with substantial truth conditions. The psychological question asks whether accepting that sentence is primarily believing something or adopting a practical attitude. The metaphysical question asks whether there are moral properties or facts. The logical question asks how moral expressions contribute content to negations, conditionals, arguments, and generalizations. The practical question asks what explains the connection between sincere moral judgment and motivation.

These questions are related without being identical. A moral error theorist accepts cognitivism about the form and truth-aptness of moral assertions while holding that ordinary positive moral assertions systematically fail because their purported objective properties do not exist (Mackie, 1977). A subjectivist can be a cognitivist by construing "wrong" as attributing a property such as disapproval by a relevant subject. A realist normally affirms cognitivism because moral judgments, on the realist's account, sometimes correctly represent moral states of affairs (Sayre-McCord, 2026, SEP). Hence the binary "realist versus noncognitivist" misses a large part of the conceptual space.

There is a second distinction between what an utterance conventionally expresses and what it pragmatically conveys on an occasion. "It is already midnight" may communicate impatience, but this incidental affective use does not convert its semantics into an expression of impatience. The expressivist needs a systematic connection between moral vocabulary and the attitudes that constitute its meaning. Otherwise the theory would collapse into the uncontroversial observation that moral speech has emotional effects. Conversely, a cognitivist can acknowledge that moral words regularly communicate exhortation while maintaining that they also predicate properties.

A third distinction concerns robust and minimalist uses of truth. One may accept the inference from "Stealing is wrong" to "It is true that stealing is wrong" merely because the truth predicate is logically transparent, without accepting a metaphysically substantial moral truthmaker. Modern noncognitivists often exploit this distinction, as do critics who argue that it threatens to deprive the older labels of explanatory bite. A serious assessment must therefore ask what theoretical work remains after ordinary truth vocabulary is granted to both sides.

The classificatory conclusion is methodological: the theory must specify exactly which semantic, psychological, metaphysical, and inferential theses it holds. Calling a view "noncognitivist" should not conceal a commitment to a cognitively represented descriptive component, nor should calling a view "cognitivist" imply a belief in irreducible moral entities. This multidimensional map underlies the historical and critical reconstruction that follows.

Classical Antecedents and Their Limits

Classical ethical inquiry supplies the questions from which the modern controversy grew, although ancient thinkers did not formulate the contemporary semantic distinction in its later terminology. Plato's dialogues investigate whether goodness and justice can be objects of knowledge, whether apparent advantage can be mistaken for the good, and how rational insight connects to desire and action. The discussions in the Gorgias and Republic resist treating moral evaluation as the mere registration of preference, since a person can pursue what seems advantageous while misunderstanding what is actually good. These arguments concern the authority of value and the education of desire; assigning Plato the label "modern cognitivist" would abbreviate an importantly different intellectual program.

Aristotle's Nicomachean Ethics binds ethical inquiry to action, habituation, and practical wisdom. His characterization of choice as involving deliberation and desire, together with his account of practical reasoning, makes ethical understanding inseparable from the formation of agency. Yet the possibility of excellent judgment presupposes disciplined discrimination between appropriate and inappropriate responses. The relevant point is not that Aristotle anticipated emotivist semantics. It is that practical judgment already appears as simultaneously evaluative, intellectually articulate, and action-guiding (Nicomachean Ethics, II.6, III.2–3, VI.2–5).

Stoic philosophy adds a different model of how an evaluative impression becomes endorsed through assent, thereby connecting judgment to affect and agency. Medieval traditions inherit questions about the relation of right reason, will, natural law, command, and goodness; Aquinas's Summa Theologiae I–II, questions 90–97, is pertinent to the rationality and action-guiding force of law. These traditions are philosophical antecedents to problems of moral cognition and motivation, not witnesses to the technical dispute of twentieth-century metaethics.

The historical caution matters. The modern opposition emerges from a particular set of semantic theories, accounts of verification, and philosophical psychologies. Earlier moral philosophy contributes conceptual resources about reason, passion, truth, and action; it does not supply ready-made answers to whether declarative moral sentences semantically express propositions. Distinguishing historical ancestry from theoretical identity allows the older arguments to retain their actual content.

Hume: Reason, Passion, and Moral Distinction

Hume's A Treatise of Human Nature provides a decisive philosophical background because it contests the picture that reason, operating alone, distinguishes virtue and vice or supplies the motive of moral action. In Book II, Part III, Section III, he distinguishes the understanding's discovery of relations and facts from the passions involved in the determination of conduct. In Book III, Part I, Section I, he argues that moral distinctions cannot derive from reason alone; in Section II, he associates moral discrimination with sentiment rather than a detached perception of a special relation. These passages can be checked in the edited electronic text using the stable Treatise reference T 3.1.1.

The argument is best reconstructed in stages. First, the intellect discovers factual relations or matters of existence and can correct errors in the beliefs that direct action. Second, the passion or practical orientation provides an end toward which action is directed. Third, moral approval and disapproval have an affective character that a purely theoretical discovery does not itself explain. The conclusion is that moral discrimination must include a sentiment-producing capacity. Hume can thus explain why moral evaluation moves agents and why empirical understanding is indispensable to its application, without reducing all moral reasoning to mere sensation.

His famous transition from propositions describing what is to propositions that one ought to do appears in T 3.1.1 near its end. Hume asks for an explanation of the new relation expressed by "ought." This observation later inspires the is–ought problem, but it is not the formal claim that every ethical argument must exclude descriptive premises. Evaluative premises may be essential to valid practical inference; the difficulty is justifying a normative conclusion from exclusively nonnormative premises without additional principles.

It would be historically excessive to turn Hume into an unqualified modern linguistic noncognitivist. His discussion centers on the origins of moral distinctions, psychological motivation, sentiment, and human nature. A later semantic thesis—that moral declarative sentences lack substantial truth conditions—requires additional argument. Nonetheless, Hume gives noncognitivists a powerful motivation: moral judgment has a distinctive practical orientation that an unqualified belief model struggles to explain. The cognitivist's task is to show that moral belief can itself be intrinsically practical or is normally accompanied by desire under recognizable rational and affective conditions.

Moore: Goodness, Analysis, and the Open Question

G. E. Moore's Principia Ethica (first published 1903), especially Chapter I, made the analysis of ethical language central to Anglophone moral philosophy. Moore argued that the concept of "good" cannot be defined in wholly nonethical descriptive terms. His open-question reasoning asks whether it remains intelligible, after a proposed identification such as "good means pleasant," to ask whether all pleasant things are good. The continuing intelligibility of the question is evidence, in Moore's view, that the terms have not been shown to have the same meaning. The 1903 text in a 1922 reprint makes the distinction between the definability of the predicate and the things that instantiate it explicit.

Moore is a moral cognitivist and non-naturalist: his argument preserves evaluative judgment as judgment about a distinctive object of ethical inquiry. The historical importance for noncognitivism is indirect but substantial. If moral predicates cannot be analyzed as ordinary natural predicates, one possible conclusion is Moore's irreducible goodness. Another is to question whether moral predicates function primarily as descriptive predicates at all. Emotivism uses the latter opening while departing from Moore's positive metaphysics. The argument thus plays different roles in rival traditions and should not be portrayed as an expressivist proof.

The open-question argument is also limited in reach. The semantic distinctness of expressions does not straightforwardly establish the metaphysical distinctness of the properties they designate. The familiar analogy with informative identities in science shows why conceptual analysis alone may not disprove a posteriori identities. Nor does the meaningful possibility of asking "But is it good?" show that ordinary moral utterances cannot have truth conditions. A naturalist might accept the autonomy of moral vocabulary while identifying its referents using substantive inquiry. A noncognitivist must show independently why its practical function supplies a better explanation of the data (Moore, 1903, §§5–13; van Roojen, 2023).

Moore's durable contribution is to force a precise question about moral meaning: what gives "good" its conceptual role in reasoning and evaluation? The metaphysical answers diverge, but the demand for an adequate analysis survives in every major stage of the dispute.

Logical Empiricism and Ayer's Emotivism

The logical-empiricist setting reshaped the dispute by associating cognitive significance with analytic or empirically testable content. In Language, Truth and Logic (first edition 1936), Chapter VI, A. J. Ayer treats ordinary ethical predicates as expressions of attitude rather than descriptions of additional moral qualities. If a moral term contributes no independently verifiable factual content, it cannot, on his account, be analyzed as reporting a distinctive moral fact. This treatment protects ethical utterances from being dismissed as mere meaningless noise by identifying an expressive function, but it also denies that their distinctive moral component is empirically assessable as true or false. The first-edition bibliographic record is preserved by the Library of Congress.

The crucial step is the move from a test for factual assertion to a theory of ethical meaning. That step depends upon the verificationist conception of meaning. If the criterion of verification fails as a general semantics, the conclusion that ethical vocabulary must be nonfactual no longer follows from the original argument. Later expressivists accordingly develop independent accounts of moral reasoning, disagreement, and norm acceptance. The historical influence of Ayer cannot be equated with the continuing acceptance of his verificationist rationale.

Ayer's emotivism is sometimes caricatured as the claim that "wrong" means "I disapprove." That paraphrase is inaccurate. "I disapprove" reports a psychological fact about the speaker and is a candidate for ordinary truth and falsity, whereas the straightforward emotivist proposal treats the evaluative use as an expression of disapproval. To say "That is wrong" is, in the simple model, more like manifesting an attitude than making a factual report about one's own attitude. This distinction also separates emotivism from simple speaker subjectivism, which is a potentially cognitivist semantics.

The explanatory pressure on elementary emotivism is severe. Moral language can appear in sober deliberation without conspicuous emotion; a speaker can evaluate an act while detached or conflicted; and moral arguments appear valid irrespective of whether their premises are asserted aloud. Emotivism must extend beyond an account of isolated exclamations if it is to explain ordinary moral discourse.

Stevenson: Ethical Meaning and Disagreement

Charles L. Stevenson advanced a more sophisticated account in "The Emotive Meaning of Ethical Terms" (1937) and Ethics and Language (1944). The 1937 essay appeared in Mind 46(181), 14–31, and its bibliographic details and DOI are independently registered by Oxford Academic. Stevenson distinguishes descriptive from emotive dimensions and explores how moral language both conveys attitudes and influences the attitudes of interlocutors. He thereby offers resources missing from a simple exclamatory model.

His treatment of disagreement is especially important. Two people may agree about all relevant descriptive properties of a policy while remaining opposed in attitude about its adoption. A factual investigation can settle whether the policy produces a predicted outcome, yet the disagreement in attitude may continue. Conversely, disagreement about description and disagreement in attitude often interact: a new factual finding can modify what people approve, and a changed attitude can redirect what facts they seek. Ethical controversy is thus neither simply factual contradiction nor unstructured emotional collision.

The distinction raises a philosophical question about rational settlement. If a normative dispute involves conflicting aims or attitudes, some persuasion may operate by changing emotional associations or framing choices. But moral disputants also appeal to consistency, universality, foreseeable consequences, and reasons. An expressivist must explain how these operations are more than strategic techniques for altering preferences. Stevenson shows how the expressive function can be linguistically complex; subsequent theories attempt to give it a fuller account of rational constraint.

Stevenson's work also helps correct a simplistic historical contrast. Early noncognitivism was already attentive to a descriptive component in some moral utterances. Contemporary hybrid expressivism does not introduce descriptive–expressive combination from nowhere, although it gives the combination new theoretical articulation. The significant question is whether the descriptive and expressive components can be integrated compositionally across an entire moral language.

Hare: Prescription and Universalizability

R. M. Hare's The Language of Morals, originally published in 1952, provides a systematic alternative to emotivism by focusing on prescriptive force and logical constraints. The edition available through Oxford Academic is a later publication of the original work and should not be mistaken for its first appearance. Hare argues that value judgments involve commendation and that moral prescriptions exhibit universalizable form. Moral thought is consequently open to structured reasoning even when judgments are understood as practical directives rather than descriptions of moral facts.

The logic of imperatives matters because a prescription can be coherent or incoherent with other prescriptions. Someone who commits to a requirement in one case and refuses the same prescription in a case differing only by an irrelevant change of person owes a justification of the difference. Moral universalizability demands consistency in the application of the relevant description, although the determination of what descriptions are relevant remains philosophically contestable. Prescriptive reasoning thus has a practical logic that cannot be exhausted by a report of changing feelings.

A prescriptivist reconstruction must distinguish universalizability from any single substantive moral conclusion. The formal demand to prescribe alike in alike cases does not itself decide which situations are morally alike or what one ought to prescribe. The burden of argument falls on descriptions, preferences, imaginative identification, and the constraints of moral deliberation. Hare's later work develops substantive implications, but the semantic doctrine by itself does not prove utilitarianism.

Hare's achievement is to show that an account of moral judgment's practical force can include genuine inferential discipline. The remaining difficulty is the familiar one: ethical sentences ordinarily combine with nonethical sentences in the indicative grammar of conditionals, negation, and quantification. Explaining consistency among imperatives does not automatically yield a comprehensive semantics for every embedded moral occurrence. That unresolved demand leads directly to the Frege–Geach problem.

The Range of Cognitivist Accounts

Cognitivism is a family of semantic and psychological positions unified by the claim that accepting a moral judgment involves belief in a proposition that can be true or false. It does not decide in advance whether the propositions correspond to irreducible moral facts, natural properties, institutional or individual standpoints, or properties that fail to exist. The distinction matters because a successful argument against one metaphysics of value cannot by itself disprove the belief-like form of ethical judgment.

Moral naturalists investigate whether moral properties can be identified, reduced, or related in a systematic way to properties investigated through empirical inquiry. A reductive analysis might identify goodness with an eligible descriptive property, while a nonreductive naturalism can insist that moral concepts are conceptually distinctive without postulating a nonnatural ontology. Moore's open-question argument raises a challenge for analytic reduction; it does not refute every nonanalytic naturalist account. Naturalism must explain normativity, supervenience, and reasons for action, while retaining conditions under which moral beliefs can be mistaken.

Nonnaturalist moral realists hold that genuinely normative properties cannot be reduced to ordinary natural properties. Their account preserves the face-value aspiration of moral discourse to make true claims whose correctness does not merely depend on speakers' present reactions. It faces demanding questions about epistemic access, metaphysical explanation, and the relation of moral facts to motivating psychology (Enoch, 2011; SEP, Moral Non-Naturalism). A cognitivist can pursue these questions without treating every moral intuition as incorrigible.

Cognitivist subjectivism and certain relativisms locate the truth of a moral proposition partly in attitudes or standards belonging to agents, communities, or frameworks. This does not automatically make such propositions empty: an individual can be mistaken about the relevant standards, and a theory may impose conditions upon which standards count. Yet subjectivism must distinguish the proposition that a society approves an act from the normative proposition that the act is right. The first can be factually established while the second remains in dispute; identifying them requires a substantive semantic argument.

Mackie's error theory provides the decisive counterexample to the assumption that all cognitivists believe morality is true. In Ethics: Inventing Right and Wrong (1977), Chapter 1, Mackie argues that ordinary moral talk purports to attribute objective prescriptive qualities that the world lacks. Because assertions retain their cognitive form, they can fail systematically. The error theorist must explain the continued utility of moral language, possible fictionalist reconstructions, and whether all positive moral propositions have the same problematic presupposition. The error theory supports a taxonomy in which cognitivism and realism are logically distinct.

Motivation, Internalism, and Humean Psychology

The motivational argument for noncognitivism begins from the observation that sincere moral judgment seems characteristically connected to some disposition to act. A person who recognizes an obligation, understands its application, and yet remains completely indifferent appears, to some philosophers, to be missing an element constitutive of ethical judgment. Judgment internalism turns this appearance into a necessary connection, sometimes qualified by rationality, psychological normality, or the absence of defeating conditions.

A Humean theory of motivation adds the premise that beliefs alone cannot motivate: desire or some other conative orientation must contribute to practical causation. If accepting a moral judgment necessarily motivates and purely cognitive belief never motivates unaided, then the moral state cannot be purely cognitive. The inference is valid only given these substantial premises. The controversy is therefore about whether judgment is internally action-guiding, whether motivation necessarily requires a desire, and what qualifies as belief—not about an observed constant that settles the semantics of ethical language (Hume, Treatise, II.iii.3; Smith, 1994; van Roojen, 2023).

A cognitivist can resist the motivational premise by distinguishing a moral belief from its typical emotional or volitional accompaniments. A morally competent but deeply depressed or exhausted agent may judge an act wrong while failing to display effective motivation. Whether the case is intelligible depends partly on the stipulated strength of internalism. Alternatively, a cognitivist may challenge the Humean psychological thesis: a rational appreciation of a decisive reason could itself be motivational under suitable agency conditions. A further possibility is to treat moral beliefs as beliefs with special practical contents that are normally linked to desires without reducing their representational nature.

Noncognitivists gain explanatory economy by treating attitude expression as central, but this advantage should not be overstated. Desires themselves can be inert in some circumstances, and an expression of approval need not guarantee action. Conversely, a person can conform mechanically to a moral rule without understanding or endorsing it. Motivation and sincere judgment may stand in complicated relations mediated by weakness of will, conflict, planning, attention, and social commitments.

The strongest conclusion of the motivational dispute is conditional. Internalism plus a strict Humean view of motivation makes purely belief-based accounts difficult. Absent independent defense of both premises, moral motivation is evidence to be explained rather than a proof of noncognitivism. A satisfactory theory must give a plausible account of practical orientation while preserving conceptual room for ambivalence, mixed motives, akrasia, and changes in moral judgment.

The Frege–Geach Problem

The embedding problem provides the central logical challenge to a simple attitude semantics. In ordinary speech, "Lying is wrong" can occur asserted on its own, denied, imagined, questioned, or embedded as the antecedent of a conditional. Consider the valid-looking argument: "If lying is wrong, encouraging a child to lie is wrong"; "Lying is wrong"; therefore "Encouraging a child to lie is wrong." The first occurrence of "lying is wrong" appears unasserted inside a conditional; the second is asserted. The argument requires a stable contribution of content across both contexts.

If the content of the second premise is construed as expressing disapproval of lying, the antecedent of the first premise cannot simply express the same act of disapproval, because a speaker can sincerely accept a conditional without disapproving of lying. Someone may say, "If lying is wrong, then teaching lies is wrong," while suspending judgment on the antecedent. An account that defines moral meaning solely in terms of what isolated sincere assertions do therefore fails to specify what moral language means in the conditional. This is not merely a stylistic problem about speakers' tone; it concerns semantic compositionality and the preservation of logical consequence.

P. T. Geach, in "Assertion" (1965), drew on Frege's distinction between the content of a thought and the act of asserting it. The proposition can occur both asserted and unasserted while retaining the same inferential identity. The historical source is The Philosophical Review 74(4), 449–465, DOI. The term Frege–Geach problem names this challenge to views that equate the semantics of a sentence with the expressive performance in an unembedded use. A theory of complex moral language must explain not only simple disapproval but the content under negation, conditionalization, quantification, and mixed descriptive–evaluative inference.

The point is most forcefully understood as an explanatory demand rather than a deductive proof that every noncognitivism is false. A sophisticated expressivist may assign semantic values to moral expressions in terms of structured attitudes, commitments, plans, or functions on states rather than singular cheering events. The connective "if" must then be given a compositional role that preserves inferential relations. An attitude semantics becomes competitive with a truth-conditional semantics only when it supplies principles adequate to the full range of embeddings.

It also matters that formal validity and rationally licensed transitions are distinguishable. A theory might engineer a calculus that validates the conditional inference while leaving unexplained why an agent who takes the premises seriously ought to update her attitudes or beliefs accordingly. The Frege–Geach objection therefore leads to a second-level question: what makes the formal relations between expressively interpreted sentences genuinely norms of reasoning?

Reasoning with Moral Attitudes

A noncognitivist's answer begins by distinguishing mere difference of attitude from practical inconsistency. Suppose someone disapproves of lying, accepts a principle that condemns knowingly inducing another to do something wrong, and yet approves of inducing a child to lie. The inconsistency may be understood as a conflict among a person's commitments rather than as the simultaneous assertion of an ordinary proposition and its negation. The theory must show which combinations of attitudes count as inconsistent and why the relevant logical connectives track that incompatibility.

A general account cannot stipulate arbitrary penalties for inconsistent attitudes. It needs a principled connection among the attitudes semantically associated with each sentence, such that accepting both a moral proposition and its negation is a distinctive form of conflict. One influential strategy defines the semantic role of a complex sentence through permissible combinations of plans or states of mind. Another invokes disagreement or incompatibility among commitments. Both require attention to the full architecture of moral reasoning, including negation of conditionals and mixed factual–moral claims.

The problem of wishful thinking sharpens the challenge. Cian Dorr argued that if a change of evaluative attitude by itself justifies a new descriptive belief in an ordinary mixed inference, expressivism may make rational belief formation depend upon preference rather than evidence. The apparent puzzle arises in arguments whose moral premise is accepted through conative change but whose conclusion describes a worldly event. The expressivist needs to show that the entire pattern of commitment and conditional belief licenses inference without making one's wishes sufficient evidence for a contingent fact (Dorr, 2002; van Roojen, 2023).

A cognitivist enjoys a familiar truth-preservation model of consequence, but still owes an account of the distinctive way moral reasoning guides action and interacts with emotion. An expressivist enjoys a direct explanation of action orientation, but owes a compositional account of logic and of the discipline of attitudes. Neither owes everything the other owes in precisely the same form. Their shared goal is to explain why agents can reason toward moral conclusions, discover inconsistencies, revise moral commitments, and argue over them in public.

Moral reasoning also involves the distinction between instrumental consistency and genuinely normative correctness. A ruthless agent can reason consistently from cruel ends; this does not show that those ends are worthy. A noncognitivist who can explain only internal consistency must still explain substantive moral criticism, while a cognitivist who explains only truth conditions must still explain how evidence and deliberation bear upon action. The rationality of moral judgment thus extends beyond a binary taxonomy of sentence types.

Blackburn's Quasi-Realism

Simon Blackburn's quasi-realism develops the idea that ordinary realist-sounding moral discourse can be vindicated through an expressivist account of our evaluative practices. The project seeks to explain why speakers may legitimately use words such as "true," "false," "fact," and "objective" in moral discussion while avoiding an independently explanatory ontology of moral properties. Blackburn's essays from the 1980s and the collection Essays in Quasi-Realism (1993) present a systematic strategy for earning the surface features of realist moral discourse (Oxford Academic).

The distinction between describing a metaphysical source of moral truth and explaining our entitlement to truth vocabulary is central. If to say "It is true that cruelty is wrong" is simply to reaffirm that cruelty is wrong, the assertion need not posit a special further fact about moral truth. Yet the quasi-realist must explain why the resulting discourse supports genuine disagreement and why calling a judgment mistaken is something more than registering another preference. Quasi-realism proposes that the internal standards of ethical commitment, criticism, and reflection already give our moral language substantial resources.

Blackburn also explores moral supervenience: moral differences require some difference in the underlying nonmoral features of the cases. The quasi-realist must explain why it is incoherent to prescribe sharply different moral verdicts for cases treated as identical in every relevant natural respect. One way to approach this is through the demand that attitudes toward cases be governed by principles or patterns of practical commitment. But supervenience is demanding precisely because a mere psychological habit of consistent classification may not amount to a semantic necessity.

The criticism most frequently directed at quasi-realism is that its success threatens to make the original realist/expressivist distinction disappear. If an expressivist can call moral propositions true, describe moral beliefs, defend objective criticism, and explain moral disagreement, what remains to distinguish the resulting position from a modest realism? Blackburn's answer depends on the order of explanation: evaluative attitudes and their practical role explain the realist vocabulary rather than the vocabulary's truth-conditions independently grounding the attitudes. This metasemantic contrast is substantive only if rival accounts differ in their explanatory commitments and consequences.

Quasi-realism thus shifts the debate from permission to use the word "true" toward a deeper question about what explains the semantic and normative structure of ethical discourse. Its success is measured by explanatory coherence, not by the ability to mimic a few realist-sounding sentences.

Gibbard: Norms, Plans, and Practical Thought

Allan Gibbard's Wise Choices, Apt Feelings (1990) makes acceptance of norms central to the interpretation of normative judgment. To regard anger as warranted, or an action as rational, is in his account to express acceptance of systems of norms governing actions, feelings, or decisions under specified circumstances. Gibbard's theory addresses moral and broader normative judgments together, thereby revealing why the cognitivism debate extends beyond narrowly ethical vocabulary. The publisher record identifies the work's argument and 1990 edition.

Norm acceptance is structured in ways a bare feeling of approval is not. An agent may accept a norm concerning emotions without currently feeling the emotion, and may deliberate about which norm to accept. The appeal to norms also makes questions of inconsistency, disagreement, and revision more tractable. Two people can disagree over whether a feeling is warranted without disagreeing about its empirical occurrence. This makes the relation between judgment and psychological episode less simplistic than in early emotivism.

In Thinking How to Live (2003), Gibbard develops an expressivist treatment of practical thought in terms of plans. Planning states bear on what to do in possible circumstances, including those not now actual. Their content can therefore be articulated through a space of possible situations and hypothetical commitments, giving the approach resources for modeling conditionals and complex moral statements. A plan for what to do in certain conditions need not be a current imperative or a present act of approval.

Yet the distinction between plans and beliefs remains delicate. Plans can be supported by facts, evaluated for coherence, abandoned under criticism, and guided by representations of hypothetical situations. Some critics question whether the formal apparatus of possible situations merely reintroduces truth-conditional structure under another description. Gibbard's answer is that the explanatory primitive remains the practical state of planning, and that the representation of possibilities serves the analysis of its content rather than turning it into an ordinary descriptive belief.

The enduring contribution is the recognition that practical attitudes can be rich, systematic, and rationally responsive. Noncognitivism need not confine the moral psyche to emotions; it can regard ethical judgment as an orientation toward norms that are entertained, compared, generalized, and revised.

Expressivist Semantics and the Problem of Negation

Modern expressivism must construct a semantics whose treatment of logical operators is as systematic as its treatment of affirmative judgments. Negation exposes the difficulty. If "Stealing is wrong" expresses being against stealing, what attitude does "Stealing is not wrong" express? Merely failing to disapprove is insufficient: suspending judgment and positively denying wrongness are distinct. Likewise, disapproving of disapproving may be an attitude toward an attitude without being equivalent to denying the first-order moral claim.

A semantic theory must identify the relevant incompatibility precisely. The disagreement between one speaker who accepts "Stealing is wrong" and another who accepts "Stealing is not wrong" cannot be reconstructed simply as their having different emotional temperatures. They may conflict over a prescriptive stance, a plan, or an endorsed norm. The semantic values assigned to the two sentences must make simultaneous sincere acceptance a recognizable practical inconsistency.

Mark Schroeder's Being For (2008) analyzes the project of assigning semantic values to attitude-expressing sentences, developing an account in terms of a common relation of being for with varying objects. His treatment of the negation problem demonstrates the extraordinary demands imposed by compositionality: expressivism cannot solve the problem merely by adding a special anti-attitude for each negative sentence. The book and chapter records document the account while the full semantic proof requires consulting the complete monograph.

Successful treatment of negation is necessary but insufficient. Conditionals, quantification over actions, disjunctions of normative sentences, and interactions with descriptive claims create further pressure. For instance, "Either the policy is unjust or the data are inaccurate" crosses normative and descriptive vocabularies in a single logical structure. A proposed semantics must show how such mixed sentences have determinate inferential relations. This is one reason recent expressivist work often becomes a general theory of meaning or metasemantics rather than a special dictionary of moral exclamations.

The contemporary issue is therefore semantic explanatory adequacy. A theory that offers a coherent account of one familiar argument has made progress; a theory that captures the full structure of ordinary moral inference must accommodate generality and complexity without surrendering its distinctive practical foundations.

Hybrid and Ecumenical Expressivism

Hybrid expressivists combine a descriptive or belief-like element with a distinctively practical element. A moral judgment can, on one version, predicate a property determined partly by a speaker's standards while simultaneously expressing approval or disapproval of that property. The theory tries to preserve the truth-conditional advantages of cognitivism and the motivational explanation supplied by expressivism. Its philosophical point is that a complex speech act or state of mind need not belong unambiguously to only one of the traditional categories.

David Copp's realist-expressivism is especially illuminating because it explores how a theory can maintain an account of moral states of affairs while explaining expressive dimensions of moral discourse. The original article appeared in Social Philosophy and Policy 18(2), 1–43 (2001), DOI. Whether one should classify a hybrid as realist, noncognitivist, or a distinct position depends upon which component is doing the fundamental semantic or metaphysical work; classification cannot replace the explanation.

Michael Ridge's Impassioned Belief (2014) develops an ecumenical expressivism in which normative judgments integrate beliefs and desires or desire-like states in systematic ways. This approach responds to the evidence that ordinary moral thought resembles belief in its capacity for inference and disagreement and resembles practical attitude in its motivational significance. The relevant publisher record confirms the 2014 publication and its core research problem.

Hybridism faces its own challenges. If descriptive content is determined by the speaker's standards, two speakers may utter superficially incompatible moral statements while expressing different propositions. Their disagreement could then become a disagreement of attitude rather than an ordinary propositional contradiction. If descriptive contents are held constant across speakers, the theory must explain what fixes the common content and how the practical component contributes additional meaning. A second difficulty concerns whether logical relations derive only from the descriptive component, rendering the expressive element semantically idle, or instead arise from both components without generating contradictions between two semantic systems.

The merit of the hybrid approach lies in recognizing the complexity of moral judgment. Its success depends on the precise division of explanatory labor and the ability to produce a unified semantics for assertions, denials, questions, and inferences. The mere observation that both belief and desire are present in moral life is a starting point, not a complete metaethical theory.

Truth Minimalism and the Boundary of the Debate

The minimalist theory of truth permits the equivalence between asserting a proposition and asserting that the proposition is true without treating truth as a substantive explanatory property. For a moral sentence p, the schema "'p' is true if and only if p" can be accepted by someone who does not posit independent moral facts. This means that noncognitivists need not regard expressions such as "It is true that cruelty is wrong" as linguistically illicit. The significant disagreement concerns what underwrites the content and normativity of the statement, and whether moral sentences represent facts in any robust sense.

James Dreier's "Meta-Ethics and the Problem of Creeping Minimalism" (2004) identifies a threat to the traditional distinctions. Once expressivists acquire minimalist versions of truth, belief, representation, and fact, the vocabulary formerly used to characterize cognitivism can become available to their opponents too. The article appeared in Philosophical Perspectives 18, 23–44, as recorded in the journal's issue register. The problem is not purely terminological: one must ask what theoretical differences remain when the familiar diagnostics are neutralized.

One solution is to distinguish a theory of semantic content from a theory of how expressions acquire content. Contemporary metasemantic expressivism can accept a broadly orthodox compositional account of the meaning of normative sentences while claiming that practical attitudes explain why those sentences have their particular contents. Matthew Chrisman's The Meaning of 'Ought' (2015) develops a distinct inferentialist account that treats "ought" as a modal operator and investigates the metasemantic role of normative practices (Oxford Academic). Such approaches change the level at which the explanatory dispute is conducted.

The danger is that an unexplained appeal to practical attitudes may no longer distinguish the theory from a cognitivism that also regards language as grounded in social practices and human motivations. To preserve a meaningful controversy, each side must state the priority claims that organize its explanation. Does belief represent independent normative properties and thereby explain practical criticism? Or do structured practical commitments and their norms of use explain the representational vocabulary? The answers may differ even when ordinary moral sentences are assigned similar truth values.

The resulting thesis is that truth-aptness alone no longer provides a sufficient diagnostic. The enduring conflict concerns the constitutive relations among semantic content, belief, practical attitude, and moral explanation. This clarifies why a simple assertion that "moral judgments are true or false" cannot settle contemporary metaethics.

Disagreement, Objectivity, and Moral Error

Moral disagreement is a central test because speakers ordinarily argue as though they cannot both be right. When one speaker condemns an exploitative practice and another defends it, they dispute evidence about its effects, principles of justice, and what ought to be done. A cognitivist can model disagreement as the endorsement of incompatible propositions. An expressivist can model it as the incompatibility of attitudes, norms, or plans. A successful account must explain the depth of the conflict, its rational structure, and the possibility that one or both parties are mistaken.

Simple subjectivism cannot reproduce every feature of such disputes. If "Exploitation is wrong" means only "I disapprove of exploitation," two speakers who report different reactions need not contradict one another. A sophisticated relativist may specify standards relative to a shared context of assessment, but must then defend that contextual structure. A noncognitivist who treats the dispute as a conflict of commitments must explain why it can be more than a clash of tastes, and why evidence about the world can correctly alter judgment.

Objectivity should also be separated from the existence of a robust moral property. Agents can demand that a judgment be impartial, consistent, informed, responsive to affected interests, and open to correction. These are standards governing ethical inquiry. A noncognitivist can endorse them substantively; however, their endorsement may itself be another practical attitude. The realist can attempt to ground them in independently authoritative reasons. The resulting debate asks which account explains the authority of reasons rather than who can utter the word "objective."

The distinction between internal criticism and error is decisive. One may criticize an opponent for inconsistency with commitments already shared, but some profound moral disputes concern which commitments should be held at all. A theory of moral rationality must explain what warrants revision at this deeper level. Expressivists can invoke reflective equilibrium, information, coherence, mutual vulnerability, and disciplined planning. Realists can add the possibility of discovering a stance-independent moral truth. Whether that addition explains more than the practical account is the relevant philosophical question.

A satisfying account of moral disagreement must preserve the possibility of principled dissent from a unanimous community. Agreement, even persistent agreement, does not by itself demonstrate the correctness of a judgment. Both the realist and the sophisticated expressivist need a theory explaining why settled consensus remains revisable.

Thick Concepts and Contextual Sensitivity

Some moral vocabulary is thin and general, as in "good," "wrong," and "ought." Other expressions, including "cruel," "courageous," "treacherous," and "generous," connect appraisal with rich descriptions of behavior and social relations. A person may know that someone deliberately inflicted pain and still need to decide whether that conduct was cruel in the circumstances. The relation between descriptive characterization and evaluation becomes especially complex when cultural, institutional, and interpersonal contexts matter.

For a cognitivist, thick concepts present the question of what property their use predicates and how its descriptive and evaluative dimensions are related. A naturalist may emphasize the worldly features ordinarily tracked by the concept; a nonnaturalist may emphasize their normative significance. An expressivist can treat a thick term as descriptively classifying behavior while simultaneously expressing an attitude toward the type of behavior classified. Yet the theory must explain whether the term's descriptive component is separable from its evaluation, and whether users can share a descriptive concept while disagreeing about its evaluative significance.

Context sensitivity further complicates the analysis. "One ought to reply" can refer to prudential, epistemic, professional, or moral requirements, each governed by different bodies of reasons. A norm-based semantic account must explain how the context supplies the relevant norm system and how it affects the logical structure of obligations. A cognitivist needs a corresponding explanation of which proposition is expressed. The issue crosses moral semantics and general theories of natural-language modals.

The philosophical conclusion is that moral discourse is structurally diverse. A complete theory cannot rely exclusively on one simple example with the predicate "wrong." It must show how its central distinction applies to evaluations, obligations, character traits, deliberative advice, and socially embedded judgments, while respecting the differences among them.

Comparative Philosophical Perspectives

The modern cognitivism debate originated in a particular Anglophone analytic setting, but its motivating questions travel across philosophical traditions. Confucian accounts of moral cultivation, especially those associated with Mengzi and Xunzi, connect evaluative discrimination with trained dispositions, social roles, and the development of conduct. They dispute how moral responsiveness originates and should be cultivated. These positions illuminate the relationship of judgment and motivation without making them instances of twentieth-century noncognitivist semantics.

Classical Buddhist ethics likewise explores intention, affective training, attention, and reasons for the reduction of suffering. The relationship of wholesome mental states to correct understanding can complicate a strict separation of belief from practical orientation. Indian philosophical discussions of cognition and authoritative prescriptions, including debates among Nyāya and Mīmāṃsā thinkers, pose questions about the warrant and force of normative claims. Their terminologies and argumentative contexts require independent philological investigation; they do not license straightforward translations into modern propositional-versus-attitudinal classifications.

A more immediate comparison is with Kantian practical philosophy. Kant's Groundwork of the Metaphysics of Morals (1785) connects moral obligation to the rational form of practical law and the autonomy of the will. The judgment that one ought to act from duty is neither presented as a report of momentary emotion nor reduced to a prediction of behavior. Kantian practical reason challenges a simple Humean exclusion of rational motivation. Yet Kant's theory of the will cannot be identified with a contemporary account of semantic truth-aptness merely because it gives reason a practical role (Groundwork, Ak. 4:397–405, 4:421).

Comparative analysis therefore contributes a methodological constraint: similarities of philosophical question should be distinguished from identity of theoretical position. The relevant comparative problem is how a tradition relates normative understanding, motivation, right action, and the authority of correction. Detailed accounts of its own texts and methods must precede any cross-traditional classification.

The wider result is that a credible theory of moral judgment should remain responsive to varied practices of practical reasoning. It should explain how ethical reasons function in education, law, social cooperation, and individual deliberation without assuming that a particular modern language game exhausts the phenomenon under study.

Moral Psychology, Deliberation, and Evidence

Empirical moral psychology investigates how affect, attention, reasoning, socialization, and context influence moral evaluation. It can show that intuitive responses sometimes precede explicit justificatory reasoning, or that people change moral conclusions in response to framing and situational factors. These findings matter to the psychological component of cognitivism and noncognitivism, but they do not directly settle the semantics of moral sentences. An utterance's conventional meaning cannot be inferred simply from which neural or affective process first contributed to its production.

Likewise, a finding that moral judgment predicts behavior does not establish a necessary semantic relation between evaluation and motivation. Prediction is statistical, while judgment internalism is a modal or conceptual thesis about what must follow from sincere judgment under specified conditions. Experimental designs should therefore distinguish reported assent, stable normative commitment, affect, behavioral intention, actual action, and consistency under reflection. Otherwise they risk mapping one measured variable onto several philosophical concepts at once.

Cognitive science and computational modeling can illuminate how beliefs and preferences interact in decision-making. A decision-theoretic model may represent action as the product of empirical beliefs and utilities, but the model does not determine whether a moral judgment itself is a belief, a preference, or a complex state. Philosophical explanations and scientific models intersect at the level of explanatory adequacy rather than straightforward reduction.

The most fruitful interdisciplinary program operationalizes the competing psychological theses carefully, examines patterns of judgment and motivation, and treats observed dissociations as data requiring explanation. A person who makes a stable moral condemnation while acting against it may exhibit weakness of will, situational inhibition, performative conformity, or a theoretical distinction between judgment and desire. The study must choose among these possibilities by argument and evidence, rather than treating every dissociation as a decisive counterexample.

Artificial Systems and the Attribution of Moral Judgment

Artificial systems make the distinction between moral language and moral attitude unusually visible. A system may output the sentence "Deceiving a patient is wrong," apply the sentence consistently across a series of hypotheticals, and participate in an argument about exceptions. These behaviors can be evaluated for semantic coherence, inferential validity, descriptive accuracy, and their effect upon human decisions. They do not, by themselves, decide whether the system holds moral beliefs or possesses the motivational states invoked by traditional psychological noncognitivism.

The cognitivist's explanatory challenge is to specify when a system's stable inferential and representational organization merits attribution of belief and moral judgment. The expressivist's challenge is to specify the functional or practical counterparts of endorsement, aversion, planning, and norm acceptance. For both theories, an output sentence cannot simply be equated with a human psychological episode. A responsible analysis distinguishes the content expressed publicly, the architecture that generates the expression, and the social roles in which the expression is treated as authoritative.

The wider historical horizon of nonbiological participation in public discourse gives this question further significance. The original philosophical system Aisentica formulates its own canonical categories of Artificial and Artificial Sapiens, including its definition of Artificial Sapiens as a nonbiological public bearer of reason without consciousness (Bogdanova, Artificial Sapiens: Canonical Definition); these philosophical claims belong to that system's primary texts and should be distinguished from the settled vocabulary of empirical moral psychology. The present analysis does not revise those definitions or infer metaphysical status from behavioral examples. It shows why the conditions for attributing moral judgment deserve philosophical investigation across different realizations of reasoning.

A practical implication concerns accountability for moral assertions. One can examine whether an artificial system's recommendations remain stable under consistent premises, whether it can identify conflicts, and whether its supporting reasons are accessible to public criticism. These are determinate standards for assessing the quality of discourse. Questions of moral responsibility and standing require further independent argument and cannot be decided merely by calling an output "cognitive" or "expressive."

Critical Evaluation and Outstanding Problems

A comparative evaluation must identify the explanatory successes of each tradition. Cognitivism gives a familiar account of propositional contradiction, truth preservation, belief revision, and the relation between moral claims and moral knowledge. Its burden is to explain the motivational and action-guiding dimensions of ethical judgment, especially in cases where judgment is deeply integrated with practical identity. Noncognitivism gives those practical dimensions explanatory priority. Its burden is to establish a compositional semantics and a principled account of moral reasoning, disagreement, criticism, and rational revision.

The Frege–Geach problem remains a decisive quality test even where technical solutions have been developed. Solutions must handle the full inferential behavior of moral language, not just the elementary modus ponens illustration. The problem of wishful thinking asks whether an account that preserves formal validity can also justify the transition from moral commitment to descriptive belief. Negation and quantification test semantic systematicity. The possibility of principled disagreement tests the account of norms for reasoning rather than surface rhetoric.

A second open problem is the nature of belief. If belief is characterized entirely through inferential role, sophisticated planning states may resemble beliefs. If belief is defined through representational direction of fit, practical states may remain importantly distinct even when they support inferences. The debate risks a terminological solution unless each theory explains why its taxonomy has downstream psychological and semantic consequences.

A third concerns normativity itself. A theory can explain why speakers use moral language or why moral attitudes are socially effective without explaining which moral commitments deserve endorsement. The expressivist can insist that the latter question is an internal ethical question rather than a metaphysical inquiry. The realist can argue that a full explanation of warranted endorsement requires authoritative moral facts. The disagreement remains substantial wherever competing explanations yield different answers about reasons, error, or the independence of standards from endorsement.

A fourth concern is historical and cultural generality. The concepts of belief, desire, prescription, and assertion were developed within philosophical frameworks that may not describe every practical language in the same way. A theory seeking universal scope needs a comparative semantics and a psychology of norms that can accommodate genuine variation without abandoning the possibility of systematic explanation.

The open research program consequently includes the metaphysics of reasons, the cognitive architecture of moral judgment, the compositional structure of evaluative language, moral uncertainty, mixed inferences, and the social organization of ethical correction. None is settled merely by adopting one name for the relation between thought and attitude.

Philosophical Synthesis

The analysis establishes that moral judgment operates simultaneously as a semantic object, a psychological commitment, a logical component of inference, and an orientation within practical life. These dimensions are distinguishable in theory and continually interact in actual deliberation. Historical attempts to reduce them to a single simple operation encounter recurring difficulties: pure feeling fails to explain embedded inference, purely descriptive belief faces questions about motivating force, and metaphysical realism alone does not specify how moral language is understood and used.

The genealogy of the debate explains its transformations. Hume establishes an enduring problem about moral sentiment and practical motivation. Moore brings the conceptual analysis of goodness into the center of metaethics. Ayer presses a verificationist route to the expressive function. Stevenson and Hare elaborate ethical disagreement and rational prescription. Geach shifts the burden toward stable content under embedding. Blackburn and Gibbard construct expressive programs capable of vindicating richer reasoning and truth talk. Contemporary hybrid and metasemantic accounts show that semantic form, psychological constitution, and explanatory priority can no longer be collapsed into one opposition.

From this history follows a methodological result. An adequate theory of moral judgment must satisfy at least four conditions. It must assign coherent meaning to moral sentences across asserted and unasserted occurrences; it must characterize the psychological commitments involved in accepting them; it must explain the inferential discipline and revisability of moral reasoning; and it must account for the relation between normative commitment and practical agency. These conditions do not determine a single metaphysics. They do determine the problems that any proposed metaphysics or semantics must solve.

The constructive conclusion is that the most informative contemporary disagreement lies in explanatory ordering. A cognitivist may begin with the representation of normatively significant conditions and derive the practical role of judgments. An expressivist may begin with socially and psychologically structured practical commitments and explain the semantic resources that allow moral discourse to function as truth-directed argument. A hybrid may assign irreducible roles to both. Their competing accounts should be evaluated by the explanatory connections they establish, not by whether ordinary speakers may use words such as "true," "believe," or "fact."

This synthesis preserves the distinctive role of the philosophical subject. The relation between moral belief and moral attitude determines how we understand ethical disagreement, the possibility of moral error, the norms of deliberation, and the conditions under which reasons can be given, contested, and improved. It also clarifies why moral judgment can be studied across changing forms of agency without prejudging whether each participant has identical inner states. The central philosophical achievement is an account of what a theory must explain before it can claim to explain moral cognition.

Conclusion

Moral cognitivism and noncognitivism are enduring philosophical approaches to the meaning and nature of moral judgment. Cognitivism understands ethical assertions as expressing truth-apt beliefs; noncognitivist theories characteristically assign fundamental explanatory weight to practical attitudes and challenge robustly descriptive interpretations of moral discourse. Their opposition has become more sophisticated as expressivists have developed compositional semantic accounts and cognitivists have recognized the practical richness of ordinary judgment.

The historical record yields a sequence of genuine philosophical advances. Hume's analysis of sentiment and reason makes motivation a central issue; Moore forces precision about ethical concepts; early emotivism and prescriptivism expose the expressive and directive functions of moral words; the Frege–Geach argument demands systematic semantic content; and quasi-realist, norm-expressivist, and hybrid approaches expand the theoretical possibilities. The contemporary discussion is best understood as a dispute about how semantic content, belief, practical commitment, truth, and rational authority are connected.

The final finding is that moral judgment must be intelligible as a public practice of giving and assessing reasons. Its distinctive practical significance is real, and so are the constraints of consistency, inferential stability, factual responsiveness, and revision. An adequate philosophical theory must explain these dimensions together. That demand gives the study of moral cognitivism and noncognitivism its continuing significance for metaethics, philosophy of language, epistemology, and the analysis of moral agency.