Lance Bush vs. Ryan Mullally — Moral Realism vs. Global Anti-Realism

A cordial 3.4-hour metaethics conversation in which a hardline anti-realist concedes a deist’s theistic metaethic is genuinely realist — and the real action is that both men wield the same constitutive/unintelligibility move toward opposite conclusions.

Date: 2026-05-25 Source: YouTube — Natural Law with Ryan Mullally — Transcript Participants: Lance Bush (global normative anti-realist; experimental-philosophy PhD, metaethics) vs. Ryan Mullally (deist / natural-law moral realist; lawyer, “Natural Law” channel) Format: Long-form cordial conversation (not a scored debate); definitions → Lance’s view → Ryan’s positive case → audience Q&A Duration: ~3h22m Result: No decisive clash. Lance concedes Ryan’s theistic metaethic is structurally realist on its own terms (while rejecting the theism). Both agree secular/naturalist realism and non-naturalist “external reasons” accounts are weak. Ryan’s retrospective: Lance’s position is a self-undermining “non-position.” Vault relevance: Lance Bush reviews Wilson, Wilson on Stance Dependence, Performative Grounding Lineage, Game Theory as Normative, Sitch vs. Mullally — Birthright & Natural Law (Mullally again, same natural-law/deist toolkit, vs. a non-lawyer host)

Context

This is the anti-realist-vs-realist pairing the vault’s metaethics cluster had been missing. Prior pages mostly orbit Andrew Wilson’s presuppositional realism; here a professional anti-realist (Bush — likely the single biggest popularizer of “stance-independent” in YouTube philosophy) meets a natural-law realist whose realism is explicitly theistic. The conversation is exploratory and gracious, which is its main structural limitation: the sharpest objections on both sides are flagged and then not pressed.

Argument structures

Lance Bush — global normative anti-realism

Realism = there is ≥1 stance-independent moral fact. A stance = any evaluative standpoint, real or hypothetical, not necessarily psychological (the Bible, or ChatGPT-generated rules, count as stances).
He is anti-realist about every normative domain — moral, epistemic, logical, prudential, aesthetic. No stance-independent reason to be logical or rational; violating logic is a practical/social faux pas (playing chess by the wrong rules), not a factual error.
Preferences are constitutive of how he analyzes “ought.” So “ought you prefer your stronger preference?” is not false but unintelligible / a loaded question — it tries to apply an “ought” from outside the only frame (“ought” is cashed out in) that gives the word content.
Universal agreement ≠ stance-independence (the French-fries analogy): even all humans who ever lived agreeing would be, at most, indirect evidence.
Practical-equivalence thesis: realists and anti-realists act identically, because everyone only ever acts on preferences; a realist complies with stance-independent facts only insofar as he subjectively values doing so. So the distinction has little practical import.
The motivation problem (his strongest blade): for a secular realist, why follow stance-independent facts that don’t align with your values? Theistic realism escapes via practical reasons (God knows best; God can enforce).
Unintelligibility thesis: non-naturalist “external reasons” (Parfit/Scanlon/Watkins) are not even false — “reason” used as an unanalyzable primitive (“a reason is what counts in favor… nothing more to say”) is a category/grammatical error, like “an animal simpliciter” or “intrinsically taller.”

Ryan Mullally — deist / natural-law realism

Bare theism: God is omnibenevolent and unchanging ⇒ “God is good” is at least one stance-independent evaluative moral fact (if God’s goodness is independent of all stances, including God’s own). Index morality to the perfect good; other indices are errors.
Natural law fits the pivotal moments: Nuremberg, abolition, civil rights, suffrage all invoked transcendent law. Realist-sounding language lines up with those moments → a comparative advantage for realism (rhetorical, if not philosophical).
Concedes the is–ought gap on naturalism: you can’t get a transcendent ought from naturalist is; Sam-Harris-style naturalist realism is weak/near-incompatible. His realism is therefore parasitic on his theism — and he brackets God’s existence for this metaethics-only frame.

Analysis — key exchanges

The “greater good” gambit fails to a constitutive move

Ryan offers “one ought prefer the greater good” (leaving good to be filled in) as a candidate stance-independent norm. Lance dissolves it: if you let the agent fill the blank, they fill it stance-dependently; pressed to “ought you prefer your stronger preference?” the question is either trivial (= “do you prefer what you prefer more?”) or unintelligible. This is a constitutive dissolution, not a refutation — and it’s the structural heart of the whole conversation (see below).

“What’s the utility of the definition?” — a good pragmatist jab

Ryan presses: if all-humans-ever-agreeing is still stance-dependent, what work does the concept do? Lance: conceptual ≠ empirical question; useful here-and-now where people disagree (like “atheism” in a hypothetically all-theist world). Decent — but it leaves a real tension: a hardline conceptual distinction with conceded minimal practical import sits awkwardly beside Lance’s own pragmatism + quietism.

Normative entanglement / pragmatic cancellation

Lance’s empirical contribution: “Do you think it’s objectively wrong to X?” fuses a normative and a metaethical question. “No” is ambiguous across three positions, one of which sounds monstrous. The needle-threading answer cancels the bad implication (“No — because nothing is objectively wrong — but it is morally wrong, repugnant, and nobody should do it”). His pilot study: that answer reads almost as well as plain “yes.” A genuinely useful diagnostic of why realists gain unearned rhetorical advantage.

Theism passes Lance’s own realism test — instantly

The surprise (Ryan’s words): Lance applies none of his famous trilemma (realist positions are false, incoherent, or trivial) to the theistic metaethic, conceding it’s realist on its own terms (false-for-Lance only because he rejects the theism). The quiet result: anti-realism’s best case is specifically against secular realism; theistic realism is the one realism that structurally survives — because it answers the motivation problem the secular realist can’t.

The deep structural observation (the gold)

Both men deploy the same weapon — the constitutive / unintelligibility move — toward opposite conclusions.

Lance: “preferences are constitutive of how I analyze ‘ought,’ therefore the realist’s question is unintelligible.” He dissolves the realist’s question by showing it presupposes a frame it then tries to step outside.
The vault’s constitutive-realism trunk (the performative-grounding lineage; Wilson’s presuppositional move) runs the identical structure to ground realism: the skeptic’s question presupposes the very norms it tries to question.

Same move, opposite targets. The Lance Bush ↔ Wilson page already flagged “the constitutive axis neither taxonomy reaches” — here Lance operates explicitly on that axis, which is why a constitutive realist and a constitutive anti-realist mostly talk past each other: each can run a parity-of-reasoning dissolution of the other (internal-critique-vs-self-defeat is the same shape). The constitutive move doesn’t settle realism; it relocates the disagreement to which frame is the inescapable one.

Where the vault stands

The vault’s metaethics (constitutive/elective, relational objectivity, stance-independence retreat) refuses the single axis this debate is fought on. It splits “stance-(in)dependent” into two questions: does it require minds? (both men: yes) and is it constitutive or elective? (the fight nobody names). Morality lives in the cell the old vocabulary has no word for — mind-dependent AND constitutive.

Reposition the debaters on that grid:

Lance is a mind-dependent electivist — morality bottoms out in preferences, agents “function fine without it,” convergence is “just biology + game theory.” That is the vault’s elective case, nearly verbatim. He calls it “anti-realism” only because the academy hands him a one-axis menu.
The vault is a mind-dependent constitutivist — morality is entailed by the structure of agency; the is–ought gap is crossed by performative grounding; convergence-under-reality-pressure is the diagnostic.

So “realism vs anti-realism” is, in vault terms, a mislabeled proxy for “constitutive vs elective.” Where the vault beats Lance’s strongest moves:

The fries move conflates categories 3 and 4. Universal love of fries ≠ objectivity — true, but fries are elective (category 4). Morality is cyanide (category 3): stance-dependent in Lance’s sense, yet convergent, error-admitting, and causally consequential. Toxicology isn’t “preferences all the way down.”
The unintelligibility thesis lands — and the vault dodges it. Parfit’s brute “external reasons” are empty; the vault agrees, because its reasons are statable — entailed by agency, not posited as primitives.
The motivation problem (“why comply?”) is the empty seat both men declared unfillable. They agreed only theism answers it. The vault answers it secularly via constitutive grounding: you can’t opt out because the norm is a precondition of your own acting; non-compliance is self-undermining, not preference-mismatch. The vault is the secular realism the debate said couldn’t exist — threaded between Ryan’s theism and Parfit’s unintelligible reasons.

The corollary: vault and Lance agree on almost everything (the binary is broken; morality is mind-dependent; category-4 is genuinely subjective; brute reasons are empty) and disagree on one near-empirical question — does biology + game theory fully explain moral convergence, or is there a residue only reality-tracking explains? The metaphysical chasm shrinks to that.

And the debate is evidence the split is well-founded — by the split’s own criterion. The framework predicts the shape of the conversation: if the standard axis is a broken proxy, debaters on it should talk past each other, keep finding “strange-bedfellows” agreement, land no decisive blow, and never locate their real disagreement — which is exactly what happens. Stronger: a hostile professional metaethicist (Bush) and the vault, working independently and reaching opposite first-order conclusions, both reject the realism/anti-realism framing as misleading. By the vault’s own Trunk-2 standard — independent convergence with no shared cause but the structure — that agreement that the axis misleads is evidence the axis really is broken. (Honest caveat: the convergence is on the negative claim only — Bush diagnoses the bad labels but stays on one axis and abandons realism; the vault adds the second axis and keeps a constitutive realism. Shared diagnosis, divergent cure.) It’s the verification-layer thesis in the metaethics domain: an accumulated framework, applied to fresh input it didn’t generate, locates the confusion the unaided reader couldn’t.

→ Operational playbook: The Constitutive Middle — Two Dialectics — how to walk this anti-realist (Bush) and this theistic realist (Mullally) to constitutive realism, step by step, with escape hatches and a quick-recall card.

Structural problems

Lance. (1) The “preferences are constitutive → your question is unintelligible” move risks unfalsifiability: a realist can run the mirror move (“your ‘it all bottoms out in preferences’ is itself a stance-dependent stipulation, not a discovered fact”). Ryan never presses this. (2) Genuine tension between hardline conceptual anti-realism and quietist+pragmatist “no substantive difference”: if realism/anti-realism make no practical difference, why be a militant anti-realist rather than a realism quietist (he is already a free-will quietist)? Ryan names this as the closing charge — the position looks like a “non-position, preferences all the way down,” which invites the self-reference worry: on what stance-independent ground is anti-realism worth arguing for?

Ryan. His realism is entirely downstream of an undischarged premise (God, bracketed for the frame). Valid as “given my metaphysic, realism follows” — but it’s a defense of theistic realism, not moral realism per se, and he concedes he can’t defend the secular/naturalist version at all. The natural-law “fits the pivotal moments” point is a rhetorical-advantage argument that he himself flags is not a truth-maker.

Toolkit (for engaging these arguments later)

Separate the two questions. When someone asks “is it objectively wrong?”, split the normative claim (“it’s wrong”) from the metaethical one (“there are stance-independent facts”). Don’t let a “no” on the second be heard as a “no” on the first (pragmatic cancellation).
Watch for the constitutive dissolution. “That question is unintelligible / loaded” is a powerful move — but note it relocates rather than settles, and the other side can usually run it in reverse. Ask: which frame is being claimed as inescapable, and why that one?
Press the secular/theistic split. Anti-realism’s motivation problem bites secular realism hardest; theistic realism answers it. Don’t defend “moral realism” in the abstract — defend the specific version whose grounding survives the motivation question.
“Can you say what it means?” Bush’s unintelligibility test (a central theoretical concept whose holder can only offer synonyms is a red flag) is a reusable scalpel — but apply it to both sides: “preference,” “stance,” and “constitutive” deserve the same demand for content as “external reason.”

Vault connections

Lance Bush reviews Wilson on Triggernometry — same anti-realist; the “constitutive axis” flagged there is operated explicitly here.
Wilson vs. Objectivism: Stance Dependence — the realist mirror of stance-dependence; relational vs. stance-independent facts.
Performative Grounding Lineage — the constitutive/presuppositional move’s 2,400-year pedigree; both debaters are doing a version of it.
The Constitutive/Elective Distinction — the vault’s home for the exact move that does the work here: Bush makes “ought” constitutive of preference (so the external question is unintelligible); the realist makes it constitutive of rationality. The distinction is the battleground.
The Stance-Independence Retreat — directly on Bush’s central term; whether “stance-independent” is doing honest work or quietly retreating.
The Four Trunks — this debate lives on trunk 1 (constitutive realism); Bush is the constitutive anti-realist mirror, which is why the move relocates rather than settles.
Moral Foundations (Wilson x Whatever) — adjacent realism dispute.
Game Theory as Normative, Not Descriptive — Bush’s practical-equivalence thesis (“norms guide behavior; everyone acts on preferences”) is the metaethical face of the same point; the convention layer does the equilibrium-selection.
Connects to the vault’s verification-layer / constitutive-realism stance: if realist and anti-realist behavior is identical, the open question “what’s doing the work?” is the same one the accumulated-state thesis answers — convergence needs a grounding, and “preferences all the way down” defers rather than dissolves it.

Open questions

Is the constitutive move symmetric? If both realist and anti-realist can dissolve the other’s question by declaring their own frame inescapable, what (if anything) breaks the tie — fruitfulness? parsimony? lived inescapability? This is the unresolved core.
Does Bush’s global anti-realism (incl. logic/epistemics) self-undermine, or is “I have no stance-independent obligation to be logical, I just prefer to be” a stable position? Ryan asserts the former; Bush treats it as obviously the latter. Neither argues it through.
Is “theistic realism survives, secular realism doesn’t” a general result, or an artifact of Bush’s particular (motivation-centered) objections?