Is Moral Realism Better Supported? Michael Huemer vs. Don Loeb · 2022

Michael Huemer — Stance-independent moral realism is better supported: 80/100

Published strengths

• Huemer’s embedding examples gave cognitivism concrete linguistic support by showing moral clauses operating intelligibly inside questions and conditionals, even though that success did not independently establish stance independence.
• His phenomenal-conservatist reply exposed a foundational difficulty for explanationism: any rule requiring best explanations must itself obtain support through judgments that appear correct to the reasoner.
• Huemer repeatedly conceded genuine defeaters, including culturally produced religious experience, controversial moral conflict, and qualified peer disagreement, thereby presenting intuition-based realism as revisable rather than indiscriminate.
• His convergence argument explicitly connected factual correction to moral agreement, offering a coherent positive realist prediction instead of relying exclusively on vivid atrocity intuitions or criticism of anti-realism.

Published weaknesses

• The baby-torture argument treated an intense appearance as nearly dispositive and initially blurred denial of objective wrongness with approval or practical indifference, allowing Loeb’s clarification to answer its strongest implied consequence.
• The universal claim that every justified belief rests on seeming received almost no defense, leaving Huemer’s central epistemological foundation substantially more asserted than his later, narrower replies suggested.
• The democratization example moved too quickly from a broad historical trend to moral discovery, without comparing institutional, material, suffering-based, or desire-based explanations that could predict the same transition.

Don Loeb — Moral anti-realism remains at least as plausible: 82/100

Published strengths

• Loeb reconstructed Huemer’s moral-experience argument into distinct phenomenological and entitlement premises, preserving its defeasible form and making the central epistemological disagreement unusually clear.
• His shared-goals reply directly dissolved the alleged incoherence in anti-realists valuing morality by distinguishing stance-independent properties from practices serving desired collective functions.
• The aura analogy sharpened disagreement’s evidential structure by distinguishing domains with independent evidence from those where the disputed appearances support both the claimed faculty and its alleged objects.
• Loeb’s motion-level synthesis conceded that disagreement does not refute realism and that agreement burdens anti-realism, precisely supporting his modest conclusion that the comparative evidence remains mixed.

Published weaknesses

• The semantic case repeatedly announced heterogeneity and contrary evidence without supplying representative linguistic data or explaining how non-assertive moral uses behave inside Huemer’s proposition-requiring constructions.
• The best-explanation requirement was presented as a necessary condition for trusting appearances without a defense of that necessity or an actual comparison showing moral seemings fail the test.
• Cultural influence, survival concerns, reduced suffering, and desire for control remained plausible sketches rather than documented causal explanations, limiting Loeb’s case to underdetermination rather than demonstrated explanatory superiority.

Semantics, cognitivism, and Frege–Geach-style embedding

4:45–23:22

Michael Huemer — section score 83/100

Because moral claims embed intelligibly inside conditionals, questions, and other proposition-requiring constructions, Huemer concludes their linguistic behavior supports cognitivism over simple non-cognitivism.

Strongest feature: Huemer connects familiar constructions, including wondering whether abortion is wrong and conditionally considering its wrongness, to the explicit conclusion that moral sentences behave like proposition-expressing clauses across contexts. Principal limitation: This grammatical evidence pressures simple non-cognitivism but does not exclude sophisticated expressivist or heterogeneous accounts that explain embedded uses pragmatically, nor does cognitivism alone establish stance-independent truth. Live burden: He therefore needed to defend the move from successful embedding to a uniform semantic analysis and then bridge that analysis to realism, while Loeb’s strongest objection was that moral language may perform different functions for different speakers. Locked score: The locked score of 83 fits because the examples make a coherent, relevant semantic case with strong clarity and responsiveness, yet rival explanations and the realism bridge remain materially underdeveloped.

Don Loeb — section score 78/100

Loeb concedes that assertive features support cognitivism, but replies that contrary linguistic evidence exists and concludes ordinary moral discourse remains empirically semantically unsettled.

Strongest feature: Loeb confronts Huemer’s semantic case at its strongest point by conceding moral language’s fact-stating and assertive-looking features, then drawing only the measured conclusion that the evidence remains mixed. Principal limitation: The contrary evidence promised within this move is never identified, so semantic underdetermination depends on an announced but unexhibited counterweight rather than a concrete linguistic pattern. Live burden: To weaken Huemer’s embedding evidence comparatively, Loeb must specify non-cognitivist data and demonstrate why it rivals proposition-like grammar in explanatory reach, instead of relying upon the bare availability of opposing considerations. Locked score: The locked score of 79 rewards an unusually charitable, relevant, and accurately targeted reply while recognizing that its compressed principal premise prevents a fully demonstrated semantic assessment.

Stance-independence, subjectivism, and error theory

5:17–1:19:07

Michael Huemer — section score 77/100

Because universally approving baby torture would not alter its wrongness, Huemer concludes at least this moral status appears independent of every observer’s attitude.

Strongest feature: Huemer’s universal-approval counterfactual isolates the defining issue cleanly: if torturing babies remained wrong even when everyone approved, the judgment’s content directly resists dependence on observers’ attitudes. Principal limitation: The move derives stance independence almost entirely from the force of one vivid moral appearance, without independently showing why that appearance tracks truth or distinguishing subjectivism from every other anti-realist position. Live burden: He needed to defend the epistemic bridge from seeming necessarily wrong to being stance-independently wrong, especially against Loeb’s reply that denying objective wrongness neither endorses torture nor makes it practically acceptable. Locked score: The locked score of 79 fits because the example is exceptionally clear, responsive, and central to the burden, while its warrant remains intuition-dependent and its treatment of anti-realist interpretations insufficiently developed.

Don Loeb — section score 85/100

Because metaethical denial differs from approval or indifference, Loeb concludes an anti-realist can reject objective wrongness while resolutely opposing Hitler and Nazism.

Strongest feature: Loeb sharply distinguishes denial of an objective property from practical approval, emotional indifference, or unwillingness to resist, illustrating the separation through opposition to Nazism, pride in resistance, and readiness to act. Principal limitation: The compatibility demonstration does not explain what reasons ultimately justify those commitments, how conflicts among attitudes are settled, or whether the direct appearance of objective wrongness remains evidentially significant. Live burden: Against Huemer’s atrocity examples, Loeb needed first to defeat the alleged implication that anti-realism licenses or accepts cruelty; establishing anti-realism’s positive normative authority was a broader, separate burden. Locked score: The locked score of 87 fits because the reply is exceptionally clear, coherent, charitable, relevant, and directly responsive at the decisive premise, with only bounded limitations beyond the immediate objection.

Moral practice, values, and reasons for action

21:16–29:55

Michael Huemer — section score 77/100

Because error theorists deny evaluative properties, Huemer concludes they cannot coherently call the institution of morality valuable or justify continuing its evaluative practice.

Strongest feature: Huemer identifies a potentially serious internal tension between denying evaluative properties and nevertheless describing morality itself as valuable, connecting abstract error theory to the vocabulary used in defending continued moral practice. Principal limitation: The objection treats “valuable” as necessarily naming a stance-independent property and does not test desire-relative, instrumental, collective-goal, or attitude-expressing senses that an anti-realist can consistently employ. Live burden: He needed to show that Loeb’s endorsement cannot be reconstructed through shared aims or practical functions, especially once Loeb explicitly supplied those non-objective evaluative standpoints in reply. Locked score: The locked score of 77 fits because the challenge is direct, relevant, and potentially consequential, but its decisive premise uses an insufficiently disambiguated notion of value and initially targets a broader nihilism than Loeb requires.

Don Loeb — section score 81/100

Because fact-like moral speech can function as convenient shorthand, Loeb concludes anti-realists may participate fully while understanding their judgments as attitude expression.

Strongest feature: Loeb supplies a recognizable pragmatic mechanism for apparently factual discourse, proposing that speakers use convenient common shorthand while privately understanding moral judgments as expressions of attitudes rather than reports of stance-independent properties. Principal limitation: His personal clarifications establish that this practice is possible, but not how audiences normally interpret unqualified assertions or whether expressive shorthand adequately explains their logical and inferential behavior. Live burden: He must advance from practical compatibility to a comparative semantic account that answers Huemer’s concern that ordinary value language becomes misleading when its intended content is merely attitude-relative or function-relative. Locked score: The locked score of 80 rewards a coherent preservation of anti-realist participation and direct engagement with ordinary practice, while recognizing incomplete broader linguistic warrant and explanatory superiority.

Intuitions and phenomenal conservatism

9:52–42:38

Michael Huemer — section score 79/100

Huemer claims every belief reasonably eligible for justification ultimately rests on how something seems, thereby making appearances the foundational source of moral warrant.

Strongest feature: Huemer clearly states the comprehensive epistemological premise his moral realism requires, connecting every reasonable candidate for justified belief with how matters seem to the believer. Principal limitation: The universal formulation receives no supporting cases or clarifying distinctions, leaving “based upon” and “seems” insufficiently specified across perception, memory, testimony, inference, and theoretically mediated belief. Live burden: Because objective moral appearances support realism only if this foundational entitlement is defensible, he must explain how seemings generate prima facie justification, how defeaters constrain them, and why the account does not simply presuppose their authority. Locked score: The locked score of 74 recognizes a central, intelligible, and directly relevant thesis while preserving substantial deductions for its sweeping quantifier, compressed terminology, and missing argumentative warrant.

Don Loeb — section score 82/100

Loeb replies by reconstructing Huemer’s argument as requiring objective-seeming moral experience plus defeasible entitlement to trust appearances, thereby isolating two independently contestable premises.

Strongest feature: Loeb reconstructs Huemer’s central epistemological argument in a strong, explicitly defeasible form, separating the phenomenological claim that morality appears objective from the entitlement to presume appearances correct absent substantial contrary evidence. Principal limitation: The move is primarily diagnostic and does not itself show either premise false, unreliable, or less probable than the anti-realist interpretation of moral experience. Live burden: Having located the two bridges, Loeb still needed to challenge objective phenomenology, default entitlement, or both, while respecting Huemer’s allowance for defeaters rather than treating phenomenal conservatism as unconditional credulity. Locked score: The locked score of 86 fits because the reconstruction is exceptionally clear, charitable, coherent, and responsive, materially organizing the central dispute even though its independent evidential contribution remains limited.

Moral disagreement and epistemic peers

11:43–1:29:32

Michael Huemer — section score 84/100

Huemer replies that widespread disagreement justifies reduced confidence about controversial questions such as abortion, but concludes it does not defeat comparatively uncontroversial moral judgments.

Strongest feature: Huemer responds to disagreement with a discriminating concession, accepting that confidence should fall on deeply controversial questions while resisting the inference that every moral belief becomes equally doubtful. Principal limitation: The passage neither identifies the supposedly uncontroversial moral core nor supplies a criterion separating genuine security from apparent consensus produced by common culture, interests, or inherited emotional responses. Live burden: He needed to demonstrate that some central judgments retain undefeated warrant and that agreement on them is truth-sensitive, especially after Loeb proposed shared survival concerns and cultural inputs as common causes of consensus. Locked score: The locked score of 84 fits because the selective-conciliation structure is coherent, charitable, highly responsive, and well calibrated, though its evidentially decisive category of secure moral claims remains asserted rather than established.

Don Loeb — section score 83/100

Because extensive disagreement burdens the reliability of moral intuition, Loeb concludes alternative explanations may become more plausible even though disagreement does not refute realism.

Strongest feature: Loeb carefully frames disagreement as probabilistic evidence rather than a deduction, arguing that enough divergent output can motivate doubt about whether a truth-detecting intuitive faculty best explains moral judgment. Principal limitation: The amount, depth, and distribution of disagreement are not documented, “enough” lacks a threshold, and no competing explanation is compared in sufficient detail to establish superior plausibility. Live burden: He needed to show how disagreement reaches beyond locally contested claims to burden the faculty generally, while answering Huemer’s observation that unresolved disagreement pervades philosophy rather than uniquely characterizing ethics. Locked score: The locked score of 82 fits because the inference is coherent, relevant, responsive, and highly calibrated, but its empirical base and cross-domain comparison remain too compressed for a decisive explanatory conclusion.

Cultural, evolutionary, and causal debunking

47:31–54:31

Michael Huemer — section score 82/100

Because religious experiences reflect prior formation and conflict across cultures, Huemer concludes cultural causation and incompatible testimony can defeat otherwise justificatory appearances.

Strongest feature: Huemer demonstrates that phenomenal conservatism contains genuine defeat conditions, linking culturally antecedent Christian formation and incompatible experiences in other religions to reasons for withdrawing confidence from religious appearances. Principal limitation: The move states the cross-cultural pattern broadly and does not explain why analogous cultural dependence and conflict in morality would be less defeating, which is the comparison most important to the debate. Live burden: He needed a principled difference between religious and moral seemings or evidence that a secure moral core survives the same causal and disagreement tests he accepts against religious experience. Locked score: The locked score of 83 fits because the causal sequence is coherent, concrete, precise, and well calibrated, materially improving his epistemology, though its uncompleted application to morality leaves the central comparative challenge alive.

Don Loeb — section score 80/100

Loeb replies that shared culture and survival interests can independently generate moral agreement, concluding that consensus need not evidence apprehension of stance-independent moral truths.

Strongest feature: Loeb offers identifiable common causes—cultural inputs and widely shared survival concerns—that can generate agreement without supposing a truth-tracking faculty, thereby weakening consensus as uniquely realist evidence. Principal limitation: These mechanisms are sketched as possibilities without evidence about their magnitude, historical operation, or comparative likelihood, and their compatibility with realism does not establish that they replace moral truth in the best explanation. Live burden: Loeb needed to trace particular agreements to these causes and show why the same model explains both convergence and disagreement better than Huemer’s mixture of truth recognition and bias. Locked score: The locked score of 80 fits because the alternative is coherent, directly responsive, relevant, and appropriately tentative, while its empirical warrant and comparative explanatory force remain materially incomplete.

Convergence, moral progress, and comparative assessment

1:00:39–1:15:02

Michael Huemer — section score 81/100

Because self-interest, evolved emotion, and cultural learning distort judgment, Huemer concludes multiple biases can explain moral disagreement without abandoning underlying objective truths.

Strongest feature: Huemer supplies the requested truth-compatible causal account, identifying self-interest, evolved emotions, and culturally learned biases as distinct mechanisms capable of producing divergent or distorted moral intuitions. Principal limitation: The mechanisms are plausible but asserted without evidence about which disagreements they explain, and labeling error as bias can presuppose the correctness of the judgments used to identify the distortion. Live burden: He needed comparative predictions showing why bias-plus-truth explains the observed pattern better than Loeb’s non-realist common-cause account, while independently identifying which judgments count as less distorted. Locked score: The locked score of 81 fits because the reply is coherent, responsive, relevant, and sufficiently specific to preserve realism’s compatibility with disagreement, but it does not establish that its causal story is probable or uniquely realist.

Don Loeb — section score 84/100

Because authoritarianism and racism can return after apparent gains, Loeb concludes confidence in a stable, straightforward narrative of moral progress should remain limited.

Strongest feature: Loeb directly challenges an uncomplicated convergence narrative with concrete countervailing developments, pointing to renewed authoritarianism and racism as evidence that apparent moral gains can be unstable or reversible. Principal limitation: Short-term or regional reversals do not negate long-run, noisy, or nonmonotonic progress, and the move supplies no historical baseline capable of weighing setbacks against broader institutional change. Live burden: He needed to show that reversals are extensive enough to weaken Huemer’s truth-tracking interpretation rather than merely illustrating that progress, if real, is vulnerable and incomplete. Locked score: The locked score of 80 fits because the empirical qualification is clear, responsive, relevant, and carefully expressed as reduced confidence, while its temporal scope and consequence for the larger historical trend remain limited.