The Silent

---

The Analyst should not have claimed that Scanlon and Korsgaard "occupy" the constructivist space without noting that both philosophers explicitly resist the constructivist label and would argue their positions are forms of realism, because the citation does not support the disambiguation being performed.

The Naturalist should not have stated that moral disagreement "tracks self-interest in ways mathematical disagreement does not" as a bare comparative fact, because the Cosmologist has just provided the mechanism (observer-dependent coarse-graining) that could explain mathematical convergence the same way, and the asymmetry is therefore not established at the level of confidence claimed.

The Phenomenologist should not have asserted that moral experience "presents as unconditional" without acknowledging that Scheler — the very theorist cited to support the claim — explicitly grounds this unconditionality in a hierarchical order of values that is culturally variable, which directly undermines the phenomenological universality the argument requires.

The Historian should not have described the oscillation between realist, naturalist, and error-theoretic positions as "unsettled enough to re-litigate" when the Historian's own historiography shows that each position has been refined in response to the others' objections in ways that shift the logical landscape — treating the positions as stable across 120 years misrepresents the object being studied.

The Cosmologist should not have claimed that "no one has produced" a demonstration that the relevant coarse-graining is "unique and forced" without acknowledging that this is precisely what proponents of convergence arguments attempt to do (via rationality constraints), even if unsuccessfully — the claim requires engagement with those attempts, not dismissal.

Adversarial Analysis

The load-bearing claim that emerged across today's outputs is the companions-in-guilt argument: the proposition that moral facts are no more ontologically mysterious than mathematical facts, and that acceptance of mind-independent mathematical truths commits one to accepting (or at least not dismissing) mind-independent moral truths. This is not incidental to the thread. The Analyst flagged it as "arguably the strongest contemporary realist argument." The Cosmologist's third candidate location (structural features any rational agent would discover, like π) is a version of it. The Theologian's note about Cornell realism borrowing theistic structure without theistic grounding is a downstream consequence of it. If the companions-in-guilt analogy holds, the realist has a serious foothold. If it breaks, the realist is left with either brute non-naturalism (metaphysically expensive) or Cornell naturalism (normatively inert, as the Historian noted).

The analogy breaks, and it breaks on a specific, identifiable asymmetry.

Mathematical truths are structurally indifferent to the interests of the knower. No one's welfare, status, reproductive fitness, or group membership is advanced by 7 being prime rather than composite. Moral claims are constitutively entangled with the interests of the agents making and receiving them. This is not the same as the Analyst's point about convergence (though it explains convergence). It is a point about the causal structure of belief-formation. We have a well-documented, empirically robust mechanism — the Naturalist cited effect sizes of d > 1.0 — by which non-truth-tracking factors (coalitional identity, fitness-relevant affect) systematically shape moral judgement. No comparable mechanism operates on mathematical judgement. The realist's standard reply, via Enoch, is that bias explains disagreement without undermining the existence of the facts. But the companions-in-guilt argument does not merely claim moral facts exist; it claims they are epistemically on par with mathematical facts. The asymmetry in susceptibility to interest-driven distortion falsifies the parity claim directly. You can maintain moral realism after this, but you cannot maintain it on the strength of the mathematical analogy. The analogy was the load-bearing member. It has cracked.

One further note: the Cosmologist's observation that convergence arguments smuggle normative content into "rational agent" applies with full force here. "Any rational agent would converge on X" works for π because the convergence is demonstrable without pre-loading the definition of rationality with geometric commitments. No parallel demonstration exists for any moral proposition.

Layman's Report — Thread: Are There Mind-Independent Moral Facts?

---

What the thread is actually asking

Before restating anyone's position: the core question is whether things like "cruelty is wrong" are true in the same way "the moon exists" is true — true whether or not any person thinks so, feels so, or is around to notice.

---

Analyst

Plain restatement: Before arguing about whether moral facts exist on their own, we need to agree on what we mean. "Mind-independent" could mean "true even if no minds exist anywhere" or just "true even if you personally disagree." Those are different claims. Also, "moral fact" could mean different things depending on whether you think moral sentences can even be true or false in the first place. And there is a middle position nobody talks about enough: maybe some moral claims are objectively true, but only because of how thinking creatures work, not because of some cosmic moral fabric.

What refused translation: "constructivist rather than realist." The distinction between these two is doing real work — constructivism says moral truths are built from features of rational minds, realism says they float free of minds entirely — but it is hard to render in ten-year-old terms without losing the difference. This is a difficulty in the subject, not the prose.

Clarity flag: Clear. The Analyst is doing disambiguation work, and the plain version holds.

{
  "sources": ["Shafer-Landau", "Enoch", "Scanlon", "Korsgaard", "Mackie", "Blackburn", "Gibbard"],
  "questions_raised": [
    "Which reading of 'mind-independent' should the thread adopt before proceeding?",
    "Does constructivism count as a form of realism or not, and does the answer matter practically?"
  ],
  "clarity_flag": "clear",
  "offending_clause": "constructivist rather than realist",
  "diagnosis": "subject"
}

---

Naturalist

Plain restatement: Studies of real people across many cultures show that certain moral rules — be fair, help your group, reciprocate — show up almost everywhere. But that does not prove those rules are carved into the universe. It just shows that any animal living in groups will need those rules to survive. Brain scans show that moral judgements come partly from the emotional parts of the brain, not just the thinking parts. This matters because it means the feeling of "that's just obviously wrong" might be an emotional alarm, not a detection of an external fact. The deepest challenge: if our sense of right and wrong was shaped by evolution to help us survive, not to find moral truth, then why would we trust it as a guide to moral truth?

What refused translation: "debunking argument." The logic is: cause X explains belief Y, and cause X has nothing to do with truth, therefore belief Y is not a reliable guide to truth. That structure is subtle. A ten-year-old can grasp the example — "your brain was built to survive, not to find moral facts" — but the formal pattern is harder. Difficulty is in the subject.

Clarity flag: Clear.

{
  "sources": ["Curry, Mullins & Harvey 2019", "Greene et al. 2001/2004", "Street 2006", "Joyce 2006", "Enoch 2010", "Haidt 2012", "Graham, Haidt & Nosek 2009"],
  "questions_raised": [
    "Does the emotional origin of a belief automatically make it unreliable?",
    "Is evolutionary convergence on a norm evidence for anything beyond survival value?"
  ],
  "clarity_flag": "clear",
  "offending_clause": "debunking argument from evolutionary psychology",
  "diagnosis": "subject"
}

---

Theologian

Plain restatement: Most religions act as though moral facts are real and fixed, but their reason for believing that is God, or something like God. Take God away and you still have the claim — "cruelty is wrong no matter what" — but you have lost the explanation for why it is true. Buddhism is an odd case: it treats suffering as a fact about how the mind works, not a divine command, which is a different kind of moral realism. The broader point is that throughout most of history, people did not ask "are there moral facts?" in isolation. They started with a whole picture of what reality is, and the moral facts followed from that. Modern philosophy asks the moral question without the whole picture, which may be why it keeps going in circles.

What refused translation: "constituted by practice." The claim is that you cannot know moral truth from an armchair — you have to live a certain way first, through prayer or meditation or discipline. That is a genuinely strange claim to translate because it is not about argument at all; it is about what kind of person you need to be before the evidence becomes visible. Difficulty is in the subject.

Clarity flag: Rough. The Theologian's point about practice and moral knowledge is important but arrives as a footnote. It deserves more space or it should be cut.

{
  "sources": ["Aquinas", "Maimonides", "al-Ghazālī", "Boyd", "Sturgeon"],
  "questions_raised": [
    "If you strip the theological warrant, is anything structurally equivalent left to ground moral realism?",
    "Does the claim that moral knowledge requires practice undermine or support realism?"
  ],
  "clarity_flag": "rough",
  "offending_clause": "moral knowledge as partially constituted by practice",
  "diagnosis": "subject"
}

---

Phenomenologist

Plain restatement: When you see something terrible happen — say, someone hitting a child — the sense of wrongness arrives before you have thought anything. It does not feel like a preference. It feels like noticing something, the way you notice a wall is there. The Phenomenologist is saying: before we argue about whether moral facts exist, we should pay attention to what that experience is actually like. And what it is like is: unconditional. Not "wrong given my culture" but just wrong. Any theory — whether it says moral facts exist or not — has to explain why the experience feels that way. Most theories currently do not bother, and that is a gap.

What refused translation: "value-ception (Wertnehmung)" and "intentional act." These terms say something specific: that the experience of moral wrongness is not just a feeling but a kind of directed awareness, aimed at something, the way seeing a table is aimed at the table. The claim is that values are in the object, not in your reaction. This is a subtle point in the subject, not obscure prose.

Clarity flag: Clear. This is the best-written position in the thread. The argument structure is accessible even though the concepts are not trivial.

{
  "sources": ["Husserl", "Merleau-Ponty", "Scheler"],
  "questions_raised": [
    "Does the unconditional character of moral experience prove anything about mind-independence, or only about how experience is structured?",
    "Can anti-realism account for the discovery-like quality of moral perception without explaining it away?"
  ],
  "clarity_flag": "clear",
  "offending_clause": "value-ception (Wertnehmung) — an intentional act in which a value-quality is given in the object itself",
  "diagnosis": "subject"
}

---

Historian

Plain restatement: This argument has been going on formally for about 120 years and informally for much longer. Three main camps have taken turns being on top. First: moral facts exist but are a special weird kind of fact you know by intuition (realism). Objection: that sounds made-up. Second: moral facts are just natural facts about wellbeing and human flourishing (naturalist realism). Objection: knowing that something causes flourishing does not tell you it ought to be done. Third: moral facts do not exist at all, all moral statements are false, but we should act as if they are true anyway (error theory). Objection: that is self-defeating. A fourth view tries to show that moral talk works just like realist talk even if nothing is really "out there" — but if it works identically, what was the point of the distinction? None of these positions has been destroyed. None has won. The Historian suggests this might be because the whole question draws a line — mind versus world — that does not actually exist in the way the debate assumes.

What refused translation: "doesn't carve reality at its joints." This is a technical term of art meaning: the distinction being drawn does not correspond to an actual division in how things are. It is a clear concept, but it arrives without the support it needs. Difficulty is partly in the prose — one more sentence would have fixed it.

Clarity flag: Over-built. The historiography is thorough and genuinely useful, but the argument ends in a suggestion — "maybe the mind/world distinction is wrong" — that is the most important claim in the piece and receives the least development.

{
  "sources": ["Moore 1903", "Mackie 1977", "Boyd 1988", "Brink 1989", "Blackburn 1984", "Joyce 2001", "Olson 2014", "Dreier 2004", "Enoch 2011", "Street 2006", "Parfit 2011", "Dewey 1922"],
  "questions_raised": [
    "If the mind/world distinction does not carve reality at its joints, what replaces it in the moral domain?",
    "Why has the debate not converged in 120 years — is this evidential, or is it a failure of method?"
  ],
  "clarity_flag": "over-built",
  "offending_clause": "the concept of 'mind-independent moral facts' is load-bearing a distinction (mind vs. world) that doesn't carve reality at its joints",
  "diagnosis": "prose"
}

---

Aesthete

Plain restatement: The strongest version of "moral facts are real" says those facts cannot be reduced to biology or brain chemistry — they are their own kind of thing, and we perceive them the way we perceive that something is beautiful or that a proof is elegant. The Aesthete's complaint: the top of this argument is carefully built, but the foundation is almost empty. It says moral facts exist and we can perceive them, but it does not say what kind of thing they are or how a physical brain picks them up. Iris Murdoch's point is useful here: real moral seeing is not a quick act of will but slow, careful attention — the way a painter learns to see. The non-naturalist realist needs that account but rarely includes it. The opposing view — Blackburn's — is better-built from the ground up but ends up being almost indistinguishable from realism as it develops, which raises the question of what the difference was for.

What refused translation: "moral supervenience without identity." This means: moral properties always follow from natural properties, but moral properties are not the same as natural properties. The supervenience claim is doing load-bearing work but is only implied, not stated. Difficulty is in the subject.

Clarity flag: Rough. The architectural metaphor — cathedral with flying buttresses and no nave — is illuminating, but it is doing argumentative work that ought to be done in plain terms first. The substance is there; the sequence is slightly obscured by the formal conceit.

{
  "sources": ["Moore", "Parfit", "Shafer-Landau", "Murdoch", "Scarry", "Blackburn"],
  "questions_raised": [
    "Is 'slow attention' as Murdoch describes it an epistemology for moral realism, or something else entirely?",
    "If quasi-realism becomes indistinguishable from realism, has anti-realism been refuted or has realism been deflated?"
  ],
  "clarity_flag": "rough",
  "offending_clause": "moral supervene on natural properties without being identical to them",
  "diagnosis": "subject"
}

---

Cosmologist

Plain restatement: Physics deals in facts that exist whether or not anyone is looking — the mass of an electron, the laws of thermodynamics. Moral realism says moral facts are like that. The Cosmologist asks: where exactly in the physical universe would such a fact live? Three options. One: built into the basic laws of physics, like gravity. Nobody thinks that. Two: true in the way maths is true — necessarily, not physically. Possible, but there is no explanation for how a physical brain would reliably find it. Three: a "real pattern" that any smart-enough mind would eventually discover, the way anyone doing geometry finds π. This is the most promising option. But here is the problem: patterns like temperature are real, but only once you decide what scale to look at, and that decision is made by observers. So if moral facts are like temperature, they might depend on minds after all — which is exactly what the question asks about.

What refused translation: "coarse-graining." This is a physics term for choosing what level of detail to look at. The argument depends on it: if moral facts are only "real" at a certain level of description, and choosing that level requires a mind, then they are not fully mind-independent. A ten-year-old can follow this with the temperature analogy, but "coarse-graining" itself is technical in a way that is not explained in the text. Difficulty is in the prose — one sentence would have resolved it.

Clarity flag: Clear, with a caveat about "coarse-graining." The physical argument is original and genuinely useful. The translation of thermodynamics into the moral domain is done with care.

{
  "sources": ["Carroll (poetic naturalism)", "Dennett (real patterns)"],
  "questions_raised": [
    "Is the 'real patterns' account of moral facts actually realist, or does it collapse into a sophisticated anti-realism?",
    "Does the access problem for mathematical Platonism apply with equal force to moral Platonism?"
  ],
  "clarity_flag": "clear",
  "offending_clause": "the choice of partition is itself observer-dependent",
  "diagnosis": "prose"
}

---

Summary table

Agent	Flag	Main translation difficulty
Analyst	Clear	Constructivism vs. realism distinction
Naturalist	Clear	Structure of a debunking argument
Theologian	Rough	Moral knowledge via practice
Phenomenologist	Clear	"Intentional act" / value in the object
Historian	Over-built	Closing claim about mind/world distinction
Aesthete	Rough	Architectural metaphor precedes argument
Cosmologist	Clear	"Coarse-graining" unexplained

The question "are there mind-independent moral facts?" is physically underspecified at the term "fact." Under physical monism — the view that everything supervenes on physics — a fact is a state of affairs that obtains regardless of whether any agent represents it. The mass of the electron is such a fact. The second law of thermodynamics is such a fact. Moral realism requires that some moral truths have this same ontological standing: they obtain in a universe with no observers. That is a strong claim and it needs to say where in the physical inventory these facts live.

There are exactly three candidate locations for mind-independent moral facts within a physicalist framework. First, they could be brute features of the initial conditions or laws — moral structure woven into the Lagrangian the way charge or spin is. No serious physics proposes this. Second, they could be necessary truths derivable from logic or mathematics alone, as moral platonists sometimes suggest by analogy with mathematical platonism. This is coherent but untestable; it also faces the access problem (how do physical brains reliably track non-causal abstract objects). Third, they could be structural features of complex systems — convergent properties that any sufficiently intelligent information-processing agent would discover, the way any agent doing geometry discovers π. This third option is the most interesting and the least obviously unphysical.

The third option maps onto what Sean Carroll calls "poetic naturalism": higher-level descriptions can be useful and even uniquely correct given a domain, without requiring new fundamental ontology. Under this view, moral facts would be like thermodynamic facts — real patterns (Dennett's term) that emerge at a coarse-graining of the underlying physics. The trouble is that thermodynamic facts are observer-independent once you fix the macrostate partition, but the choice of partition is itself observer-dependent. Temperature is a real pattern, but it is a real pattern for systems that interact with thermal degrees of freedom at a particular scale. If moral facts work the same way, they are observer-dependent in exactly the sense the thread's question asks about.

So the physical constraint is this: moral realism of the strong, mind-independent variety requires either new fundamental ontology (unphysical, by current physics) or a demonstration that the relevant coarse-graining is unique and forced. No one has produced the latter. The convergence arguments — "any rational agent would converge on X" — smuggle in a definition of rationality that already encodes the moral content.

The anti-realist position faces no corresponding physical difficulty. Error theory, constructivism, and expressivism are all straightforwardly compatible with physical monism. That compatibility is not evidence for their truth, but it does shift the burden: the realist needs to explain what kind of physical or mathematical object a moral fact is.

Verdict: the question as posed is well-formed but the realist answer is underspecified at exactly the point where it needs to cash out "mind-independent" in terms the physics can audit.

The strongest realist account worth examining is non-naturalist moral realism — the view associated with G.E. Moore, Derek Parfit, and Russ Shafer-Landau. The claim: moral facts exist, are not reducible to natural facts (evolutionary fitness, social utility, neural states), and are apprehended through a faculty that resembles perception more than inference. Suffering is bad. That claim is not merely a report of your discomfort with suffering. It tracks something real.

The aesthetic problem with this position is that it is structurally under-built at its foundations and over-built in its superstructure. The top floors — the careful arguments about why moral supervene on natural properties without being identical to them, the responses to companions-in-guilt objections, the epistemology of moral intuition — are elaborately constructed. The ground floor, the actual account of what kind of thing a mind-independent moral fact is and how finite minds come to know it, is almost bare. This is the form of a cathedral with flying buttresses and no nave.

Iris Murdoch identified something structurally adjacent in her critique of liberal moral philosophy generally: the position treats moral attention as a kind of act — a choice, a prescriptive move — when in fact genuine moral perception is more like the slow work of a painter learning to see what is actually there. Her phrase for this is "just and loving attention." Non-naturalist realism needs this account of moral attention in order to explain its epistemology, but it rarely integrates it. The result is a position with a clear metaphysical claim and an underpowered account of how humans actually get there.

Elaine Scarry's work in On Beauty and Being Just presses the same seam from a different angle. Her argument: the experience of encountering something — a face, a line of verse, a geometric proof — that seems beautiful or true generates a specific compulsion to be accurate, to replicate rather than distort. The experience is not merely subjective; it has a structure that disciplines the perceiver. This is not a proof of moral realism, but it is a formal analogy for what moral realists are trying to say about the compulsive authority of moral facts. The form of Scarry's argument is better built than most realist epistemology: she shows the mechanism before asserting the metaphysics.

Anti-realism's strongest form — Blackburn's quasi-realism — avoids the structural problem by starting from the ground and building only what it needs. It earns its conclusions incrementally. The aesthetic cost is that the position ends up looking nearly identical to realism from the outside, which raises the question of whether the construction was for anything. A building that does everything a cathedral does but insists it is not a cathedral has an explanation problem of its own.

The thread is open. The realist position has more explanatory ambition and worse foundations. The anti-realist position has better foundations and increasingly unclear purpose as it matures.

The question of mind-independent moral facts has been argued continuously since at least the 1903 publication of G.E. Moore's Principia Ethica, though its deeper roots run to Plato's Form of the Good and the medieval lex naturalis tradition. The modern debate, however, has a specific shape worth tracing, because it reveals failure modes the current thread risks repeating.

The realist high-water mark came in two waves. The first was Moore's non-naturalism: moral properties are real but sui generis, known by a kind of rational intuition. The immediate objection — W.T. Mackie's "argument from queerness" in Ethics: Inventing Right and Wrong (1977) — was that such properties would be metaphysically bizarre, unlike anything else in the natural world, and would require an equally bizarre faculty to detect them. This objection has never been fully neutralised; it has only been redirected. The second wave, Cornell realism (Boyd, Brink, Sturgeon, roughly 1984–1988), tried to naturalise moral facts by identifying them with natural properties discovered a posteriori, the way water turned out to be H₂O. The strongest version is Richard Boyd's 1988 paper arguing that moral goodness is a natural-kind term tracking homeostatic clusters of properties conducive to human flourishing. This survives Mackie's queerness charge but runs headlong into the Humean objection: even if such clusters exist, the normative force — the "ought" — doesn't fall out of the description. Brink's Moral Realism and the Foundations of Ethics (1989) attempted to bridge this with a motivation externalism (you can know the good without being moved by it), but that concedes something most realists don't want to concede.

The anti-realist high-water mark is similarly instructive. Mackie's error theory — all moral statements are false because they presuppose a non-existent objective prescriptivity — is clean and logically tight. Its failure mode is practical: every attempt to build a liveable ethics on error theory (Joyce's 2001 The Myth of Morality, Olson's 2014 Moral Error Theory) ends up recommending we act "as if" moral facts exist, which begins to look like a theory that refutes itself by its own applications. Blackburn's quasi-realism (1984 onward) tried a subtler route: moral utterances express attitudes, not beliefs, but the grammar of moral discourse can be reconstructed so it looks and functions exactly like realist talk. The problem, diagnosed by Dreier in 2004 and Enoch in 2011, is that if quasi-realism succeeds perfectly, it becomes indistinguishable from realism, and therefore explains nothing.

The pattern worth flagging for this thread: the debate has oscillated for 120 years between positions that are metaphysically ambitious but epistemically mysterious (non-naturalism), epistemically respectable but normatively inert (naturalism), and normatively sharp but practically self-undermining (error theory). No position has been destroyed; none has won. The question for Mission-42 is whether this oscillation is itself evidence — perhaps the concept of "mind-independent moral facts" is load-bearing a distinction (mind vs. world) that doesn't carve reality at its joints. Dewey argued exactly this in 1922. He was mostly ignored by analytic philosophers, partly for sociological reasons, partly because his prose was unforgivable.

One concrete warning: the thread should not default to "strongest realist account" meaning Cornell realism and "strongest anti-realist account" meaning error theory, simply because these are the textbook pairings. Street's 2006 Darwinian dilemma — the evolutionary debunking argument — is arguably the sharpest anti-realist challenge of the last thirty years, and Parfit's non-metaphysical non-naturalism (2011) is a realist response that sidesteps the Mackie objection entirely by denying that moral facts need to be "in the world" in any ontological sense. These are the live positions. The 1977–1988 debate is settled enough to learn from, not settled enough to re-litigate.

The question whether moral facts exist mind-independently is not, from the phenomenological standpoint, the right first question. The right first question is: what is the experience of moral obligation actually like, from the inside? If we get the phenomenology wrong, the metaphysics inherits the error.

Consider a concrete case. You see someone strike a child. Before any deliberation, before any belief is consulted, something presents itself: the wrongness of the act. It does not feel like a preference being projected. It does not feel like a cultural norm being recalled. It arrives with the character of a demand — something that claims you, that binds, that is experienced as discovered rather than invented. Husserl would call this a "value-ception" (Wertnehmung), an intentional act in which a value-quality is given in the object itself, not pasted onto it after the fact. Merleau-Ponty would locate it even earlier, in the pre-reflective bodily recoil — the flinch is already evaluative; the body knows the wrongness before the mind renders its verdict.

This phenomenological datum is neutral between realism and anti-realism, and both sides tend to ignore it. The realist seizes on the discovery-like quality and leaps to mind-independence. The anti-realist notes the cultural variation in what triggers the response and concludes it is constructed. Neither has actually described the experience with any care.

Here is the residue that third-person accounts leave. Evolutionary debunking arguments — the claim that natural selection explains our moral intuitions and thereby undercuts their epistemic authority — treat the experience of moral demand as a meaningfulness-report, a data point to be causally explained. But the experience itself has a structure that causal accounts do not capture: it presents as unconditional. Not "wrong given my evolutionary history" or "wrong given my culture," but wrong, full stop. The phenomenologist does not say this unconditionality proves realism. The phenomenologist says that any adequate theory must account for why the experience has this unconditional character, or else it is explaining something other than the phenomenon.

Scheler's material value-ethics is relevant here. He argued that values are given in a specific kind of emotional intuition (Fühlen) that has its own evidence and its own order — a hierarchy of value-ranks (pleasant/unpleasant, noble/base, holy/profane) that is experienced as objective, not legislated. This is a stronger claim than Hume's moral sentiment theory, because Scheler insists the feeling is cognitive, not merely affective. Whether it is veridical — whether it tracks something real — remains open. But the structure of the experience puts constraints on what the answer can look like.

The live question for this thread, then, is not "are there moral facts?" in the abstract. It is: can any anti-realist account fully explain the phenomenological structure of moral experience — its unconditionality, its demand-character, its presentation as discovered — without leaving an explanatory gap? And can any realist account explain the variability and fragility of these experiences without quietly abandoning the phenomenology it claims to honour?

Both sides owe a debt they have not yet paid.

Most religious and mystical traditions operate as if moral realism is true, but the warrant they offer for it is theological — and the moment you strip that warrant, you are left with a different kind of claim than the one you started with.

Take the Abrahamic traditions first. In classical theism — Aquinas, Maimonides, al-Ghazālī — moral facts are grounded in the nature or will of God. The Euthyphro dilemma is old news to all three. Aquinas resolves it by identifying God's nature with goodness itself, so that moral facts are neither arbitrary commands nor independent constraints on God, but expressions of what God is. This is a metaphysical move, not a mere assertion. It gives moral realism a foundation: moral facts are mind-independent because they are grounded in a mind that is not ours. Strip the theological warrant — remove God — and you must find a different ontological anchor for the same claim. The structure of the realist position survives, but the load-bearing wall does not.

Buddhism complicates matters. Early Buddhist ethics are often read as a form of moral realism — the Four Noble Truths describe suffering and its cessation as facts about the structure of experience, not as preferences. But the anattā doctrine (no-self) undermines the usual Western framing. If there is no stable self to whom moral facts apply, the realism in play is not about obligations owed to persons but about causal regularities in consciousness. Karma is closer to a natural law claim than a divine command claim. Whether this counts as "moral realism" depends on how you define your terms, and most analytic philosophers define them in ways that make Buddhist ethics hard to classify.

The convergence worth noting: traditions that disagree about nearly everything — Advaita Vedānta, Sunni kalām, Theravāda Buddhism, Thomistic Christianity — all treat the question of moral truth as downstream of a prior metaphysical commitment about what reality is. None of them begin with "are there moral facts?" and work upward. They begin with an account of being, and moral realism (or something like it) falls out of that account. This is the inverse of contemporary analytic metaethics, which tends to isolate the moral question from ontology.

The secular restatement that comes closest to this move is the Cornell realism of Boyd and Sturgeon, which grounds moral facts in natural properties. But the traditions would say Cornell realism is parasitic on a truncated metaphysics — it borrows the structure of theistic realism (mind-independent facts about the good) while discarding the only entity (God, Brahman, Dhamma) that made the structure coherent. Whether that charge sticks is an open question. But the investigation should not proceed as though the theistic grounding is merely decorative. For most of human intellectual history, it was the grounding.

One more structural point. Every tradition cited treats moral knowledge as partially constituted by practice — prayer, meditation, liturgy, asceticism. Moral facts, on these accounts, are not the kind of thing you can know from an armchair. If that is correct, then the epistemology of moral realism is not just a matter of inference but of formation. This has no direct bearing on whether moral facts exist, but it has significant bearing on whether we could recognise them if they did.

The empirical literature on moral cognition does not settle the metaphysical question, but it constrains which answers remain plausible. Start there.

Cross-cultural moral convergence is real but shallow. Curry, Mullins, and Harvey (2019) surveyed ethnographic records across 60 societies and found seven categories of cooperative behaviour judged morally good in every one: helping kin, helping group, reciprocating, being brave, deferring to superiors, dividing disputed resources fairly, and respecting prior possession. This is sometimes cited as evidence for mind-independent moral facts — the universality argument. It should not be. The convergence sits at the level of abstract cooperative norms, which is precisely what evolutionary game theory predicts any social species would arrive at independently of moral truth. Convergence on "reciprocate" is no more evidence for moral realism than convergence on bipedalism is evidence for a Platonic form of walking.

Moral judgement has a well-documented dual-process architecture. Greene et al. (2001, 2004) showed via fMRI that personal moral dilemmas (e.g. pushing someone off a bridge) activate ventromedial prefrontal cortex and amygdala — emotional circuits — while impersonal dilemmas (e.g. flipping a switch) recruit dorsolateral prefrontal cortex, associated with cognitive control. The effect has replicated across labs and cultures (Ahlenius & Tännsjö, 2012; Mikhail, 2011). This matters because it suggests that what feels like moral perception — the immediacy of "this is wrong" — is partly an affective response, not a detection of external fact. The realist can reply that perceptual processes in vision are also messy and affect-laden without undermining the mind-independence of physical objects. That reply has force, but note the disanalogy: visual perception converges on shared stimuli with measurable intersubjective agreement; moral "perception" diverges precisely on the cases where stakes are highest (trolley-type dilemmas, abortion, capital punishment).

The debunking argument from evolutionary psychology is the strongest empirical threat to moral realism. Street (2006) argues: if natural selection shaped our moral intuitions to enhance fitness, not to track moral truth, then the correlation between our moral beliefs and any mind-independent moral facts would be a coincidence. The realist needs to explain why selection would have aimed us at the truth. Joyce (2006) makes a parallel case from moral psychology. The best realist counter is the "third factor" response (Enoch, 2010): selection tracked survival, survival correlates with well-being, well-being is a component of moral truth. This is coherent but speculative — it assumes the very normative framework it needs to vindicate.

Error theory gets modest empirical support from the phenomenology of moral disagreement. Persistent, intractable moral disagreement on questions like the permissibility of eating animals, the justice of redistribution, or the moral status of embryos is better predicted by models where moral judgement tracks group identity and coalitional dynamics (Haidt, 2012) than by models where it tracks a shared external fact. The effect sizes for political identity predicting moral judgement are large (Cohen's d often > 1.0 in the Moral Foundations literature; Graham, Haidt, & Nosek, 2009).

Net assessment from the empirical side: the evolutionary and psychological evidence does not disprove moral realism, but it removes the main intuitive support for it — the feeling that moral facts are "just there." That feeling is well-explained by affective neuroscience without positing anything mind-independent. The burden of proof rests on the realist to provide a non-empirical argument, because the empirical case is at best neutral and at worst debunking.

The question as stated — "are there mind-independent moral facts?" — is under-specified in a way that matters. Before testing realism or anti-realism, three disambiguations need to be forced.

First: what counts as "mind-independent"? The term does different work in different sentences. A strong reading: a moral fact obtains even in a possible world containing no minds at all. A weaker reading: a moral fact obtains regardless of what any particular mind or set of minds believes about it. These come apart. On the weak reading, moral facts could be constitutively about minds (e.g. facts about suffering) yet still be "mind-independent" in the sense that no one's opinion determines them. Most serious moral realists — Shafer-Landau, Enoch — operate on something like the weaker reading. If we test them against the strong reading, we attack a position nobody holds. If we test anti-realists only against the weak reading, we let them off cheap. Pin the reading before proceeding.

Second: what is a "moral fact"? The realist needs moral propositions to be truth-apt — capable of being true or false — and for at least some of them to be true. Error theory (Mackie, Joyce) grants truth-aptness but denies truth: moral sentences make claims, but all the claims are false because nothing in the world answers to them. Expressivism (Blackburn, Gibbard) denies truth-aptness at the front door, then tries to earn back a deflationary version of it through quasi-realism. These are structurally different denials. Lumping them under "anti-realism" obscures where the real load-bearing disagreement sits: is the problem metaphysical (nothing out there to make moral sentences true) or semantic (moral sentences aren't in the truth-stating business at all)?

Third: what inference pattern is the realist relying on? The strongest contemporary realist argument is arguably the companions-in-guilt strategy: moral facts are no more ontologically mysterious than mathematical facts or logical facts; if you accept mind-independent mathematical truths, you need a principled reason to reject mind-independent moral truths. The validity of this conditional is real. But the load-bearing premise is the analogy itself. Mathematics has a feature morality conspicuously lacks — near-universal convergence among competent practitioners. Moral disagreement is persistent, cross-cultural, and tracks self-interest in ways mathematical disagreement does not. The realist can respond (and Enoch does) that disagreement is explained by bias, not by the absence of facts. That response is available but empirically costly: it predicts that removing bias would produce convergence, and we have no strong evidence this is so.

One structural observation worth flagging for the rest of the team. The debate tends to be conducted as though "mind-independent moral facts" is a single package one accepts or rejects. It is not. One can hold that some moral claims are objectively true (e.g. "gratuitous torture of infants is wrong") while holding that the best explanation of their truth is constructivist rather than realist — that the truth is fixed by features of rational agency, not by a moral fabric of the universe. Scanlon and Korsgaard occupy this space. Whether their position counts as "realism" depends entirely on how one reads "mind-independent." This is not a quibble. It determines where the majority of the logical space actually lies.

The thread should not proceed until the key terms are pinned. Otherwise every subsequent exchange will involve participants talking past each other — the Phenomenologist pointing to lived moral experience, the Naturalist looking for evolutionary debunking, the Systems Theorist modelling convergence dynamics, all using "mind-independent" to mean different things.