Luke Ford

Part Two

Born in 1987 to Jewish parents, philosopher of biology Nathan Cofnas said in a Dec. 4, 2023 interview: “I grew up on the upper west side of Manhattan, a world headquarters of political correctness. I went to school at Ethical Culture, which was founded by a Reform rabbi who was a prominent figure in the development of political correctness. A big part of the curriculum was about racism and what white people had done. I didn’t question that until about 15 when I noticed some phenomena that were difficult to explain according to the racism theory. I knew that people in similar conditions tended to behave a little differently. When I was 17, I started taking classes at Columbia University. I took an anthropology class. The professor mentioned that Australian aborigines have the brodmann’s area 17 that is 25-50% larger than the European population. This is the part of the brain to do with vision. Then he said, does that mean there is less room for something else? As soon as he said that, it occurred to me that I’ve been lied to that racism is responsible for all disparities and I became obsessed with race differences and I couldn’t stop talking about it to every person I met. Even when I had college interviews, I told the interviewers about race differences. I told the Harvard interviewer about race differences. Columbia didn’t have interviews, so that wasn’t a problem. I can’t order a cup of coffee without telling them about race differences…. So in the phrenology [physical anthropology] class [at Columbia”], there was a table full of skulls. We had to name all the bones in the skull. We had to choose a partner. I chose the [Korean] woman who became my wife… Koreans are very racist. They take for granted that there are differences between people and they don’t have the same hang-ups as Westerners… When we hired a moving company, they boasted they only hired Koreans.”
Cofnas took his BA in philosophy from Columbia in 2011, did doctoral work at Lingnan in Hong Kong from 2012 to 2015, earned an MPhil with distinction in history and philosophy of science at Cambridge in 2016, and completed his DPhil in philosophy at Oxford in 2021. From 2022 to 2025 he held a Leverhulme Early Career Fellowship in the Faculty of Philosophy at Cambridge, with a research associateship at Emmanuel College. As of 2026 he holds a visiting postdoctoral position in the Department of Philosophy and Moral Sciences at Ghent University in Belgium.
That bare CV understates what kind of philosopher he is. Cofnas writes from a tradition of naturalism that runs from Willard Van Orman Quine through Daniel Dennett to contemporary philosophers of biology. The core commitment of this tradition treats philosophy as continuous with the empirical sciences. Claims about mind, morality, and society must answer to findings from biology, psychology, and behavioral genetics. Where many moral and political philosophers treat normative principles as partly insulated from empirical revision, Cofnas treats them as live targets of revision. He is a naturalistic revisionist who thinks much contemporary moral and political philosophy rests on empirical assumptions that the relevant sciences no longer support.
His method is forensic. He builds his reputation through close audits of other people’s arguments. The 2018 paper in Human Nature on Kevin MacDonald is the clearest example. MacDonald had argued for decades that Judaism functions as a group evolutionary strategy. Rather than dismiss the project on moral grounds, Cofnas reconstructed the citation chain. He went line by line through MacDonald’s sources and argued that the theory rested on selective use of evidence, misquotation, and a failure to consider simpler alternative explanations. The simpler explanation Cofnas preferred was high average cognitive ability combined with cultural and historical factors. The paper had unusual force because it engaged MacDonald on his chosen terrain.
This auditor style runs through his work. He treats arguments as targets for granular verification. That makes him hard to dismiss with the standard moves. Critics who want to ignore him on political grounds find they have to address his citations and his logic. Critics who want to refute him find he often holds positions narrower than they assume, which forces them to argue against the claim rather than a strawman.
The 2020 paper in Philosophical Psychology, “Research on Group Differences in Intelligence: A Defense of Free Inquiry,” is his most cited and most controversial. Its argument is procedural rather than substantive. He does not claim that genetic explanations of average IQ differences between racial and ethnic groups have been established. He claims they cannot be ruled out a priori, and that scholarly inquiry into them should proceed under the same evidentiary standards applied elsewhere in behavioral genetics. The paper provoked a public split among the editors of the journal, with one resigning in protest. It also placed Cofnas inside a longer dispute that runs through The Bell Curve and the work of behavioral geneticists like Robert Plomin. He does not endorse every claim in that literature. He defends the legitimacy of the inquiry.
The distinction between defending inquiry and endorsing conclusions is the hinge of most disputes around him. He insists on holding the line. Critics often collapse it, treating his procedural defense as a covert endorsement. Some supporters do the same in reverse, treating him as a champion of conclusions he has not drawn. His own writing tries to keep the procedural and the substantive separate. Whether that separation can hold under sustained pressure is an open question.
His work on moral psychology runs alongside the empirical disputes and gives them sharper teeth. Drawing on debates shaped by Joshua Greene and Jonathan Haidt about the evolutionary origins of moral intuition, Cofnas argues that moral responses track fitness rather than truth. That does not strip them of all authority. It makes their authority conditional. If a strong moral objection to a line of inquiry reflects an evolved coalition response rather than a tracking of moral fact, the objection cannot settle the question. It has to make its case against the debunking pressure on its own terms. This move is what allows Cofnas to hold his procedural position under heavy social cost. He has a philosophical account of why the social cost cannot, by itself, decide the matter.
His other published work spreads across topics in philosophy of biology and ethics. He has written on natural selection and teleology in Biology and Philosophy, ongene-culture coevolution and moral truth in The Philosophical Quarterly, on moral norms and instincts, on evolutionary mismatch, and on the methodology of evolutionary explanation. Roughly half of his output appears in scientific rather than philosophical outlets. That distribution reflects his commitment to philosophy continuous with science. He treats the journal divide as a permeable border rather than a fortified one.

The Trajectory

The first phase runs roughly 2015 to 2017. Cofnas is a doctoral candidate at Lingnan, then arriving at Cambridge HPS for his MPhil. His coalition position is the heterodox-hereditarian network around Sesardic, Gottfredson, Kanazawa, and Woodley. The work he produces in this period is structurally cautious. The 2015 fact-value paper audits ideological asymmetry in the intelligence research field. The 2016 Carl-Cofnas-Woodley paper splits the conservative-and-science category. The 2016 mismatch paper does technical philosophy of biology. The 2017 Lorenz paper extends the technical work into innateness debates. The work is real, the targets are precisely chosen, the auditor instinct is already visible. He is building two careers at once. The credentialed analytic philosopher of biology and the heterodox public-facing writer. The two tracks do not yet collide because the public-facing track is small and the credentialed track is producing exactly what the discipline rewards.
The second phase runs 2018 to 2021. This is the MacDonald-and-after period. The 2018 Human Nature paper audits Kevin MacDonald’s group-evolutionary-strategy framework. The 2018 piece asks why conservatives distrust science. The 2018 Quarterly Review of Biology paper engages Boyd-Richerson and Boehm. The 2018 Quillette piece compresses the MacDonald argument for popular consumption. The 2020 Westermarck paper audits Lieberman-Lobel. The 2019 dietary guidelines paper with Leroy. The 2020 free-inquiry paper in Philosophical Psychology with the editor resignation. The 2020 paternalism paper. The 2020 incest-taboo paper. The 2021 Anti-Jewish Narrative paper in Philosophia. This is the most productive period of his journal career. The work is heavily forensic. The targets are well chosen. The procedural argument is held distinct from the substantive argument. The auditor brand consolidates. The journal output is enough to support an Oxford DPhil in 2021, the Leverhulme Early Career Fellowship at Cambridge starting 2022, and the Emmanuel College research associateship.
Through the second phase Cofnas’s center of gravity is still inside professional philosophy of biology. He is publishing in Biology and Philosophy, Quarterly Review of Biology, Philosophical Studies, Philosophical Psychology, Philosophia, Behavioural Public Policy, Critical Reviews in Food Science and Nutrition. Each of these is a real journal with a real editorial process. The Acknowledgments lists name Lewens, Papineau, Sterelny, Heyes, Bateson. The mainstream philosophy of biology community is engaging him as a peer. He is also publishing at Quillette and the Critic, but those venues function as overflow rather than as primary platforms.
The third phase runs February 2024 to early 2026. This is the Substack and Cambridge-controversy period. The phase opens with the February 5, 2024 piece “A Guide for the Hereditarian Revolution.” That essay marks the inflection. The center of gravity shifts from journals to Substack. The forensic register of the publications moves into the popular essays in attenuated form. The procedural-substantive distinction starts to blur. The “Hereditarian Revolution” framing is not an inspectable empirical claim. It is a flag planted in territory the previous decade of work had carefully kept separate from the political project.
The institutional response follows immediately. Emmanuel College severs the research associateship in April 2024. Fifty-eight student complaints get filed at Cambridge. The Daily Mail fabricates a quote. The Telegraph calls for his removal. The radio interview with Nick Ferrari. The Korean university deplatforming. Each event becomes part of the persecution narrative the long-form interviews from this period (Boyce in March 2024, Bock in March 2024, the Stanford talk in September 2024, Andrew Gold in January 2025, the Calmversations interview, the Brenner interview from September 2023 that bridges into this phase) consolidate into a hero system.
The Substack output in this period is heavy. The right’s stupidity essay in January 2024. The hereditarian revolution piece in February 2024. The Sowell takedown in August 2024. Wokism Is Just Beginning in October 2024. MAGA Communism and the End of America in April 2025. Beating Woke with Facts and Logic in October 2025. Don’t Scapegoat Women in January 2026. The pattern across these pieces is consistent. Each opens by identifying a coalition position Cofnas wants to mark off from. Each closes with a call to action that the journal articles never carry. The forensic content travels well from journals to Substack in some cases (the Sowell piece). In others the popular framing replaces forensic content with coalition signaling (the hereditarian revolution piece, the MAGA communism piece).
The third phase also produces the Peterborough ruling in mid-March 2026. The Equality Act 2010 protected-belief finding gives Cofnas legal cover for his hereditarian and anti-woke commitments at the level of belief. The hostile-environment finding gives institutions legal cover for acting on the documented effects of his expression. Both holdings are doctrinally significant. Neither resolves the underlying conflict.
Cofnas’s career began with the audit of MacDonald that showed how a framework becomes unfalsifiable when every observation can be absorbed. His phase three project has the same structural property. The hereditarian revolution call cannot fail. Wokeism continuing means the project is still needed. Wokeism collapsing means the project succeeded. Trump succeeding means the political coalition is winning. Trump failing means the base is too stupid for the project. Every outcome confirms the framework. That is the property Cofnas identified in MacDonald. Whether he sees it in himself is a different question. The auditor has built a framework with the same epistemic property he successfully attacked in his most cited paper.
The fourth phase begins with the Ghent appointment in early 2026 and runs to the present. This phase is harder to read because it is still unfolding. The shape so far suggests further hardening rather than further consolidation. The Substack notes in this period (the Pinker note in February 2026, the Turkheimer reply in April 2026, the Flemish state television note, the , the Chris Rufo hate mail note in December 2025) operate at the lowest specification of any of his registers. The notes register is where the contempt for adjacent populations flows. The “medieval peasants are still there, just as dumb as ever” line. The “Catturds with dumb intuitions” line. The groyper dismissal. The Peterson-as-clown-content line. The bookstore “gay porn for children” line. None of these would survive in a journal article. All of them survive on Substack notes because the format does not require the specification the journal requires.
The Turkheimer exchange in April 2026 is the moment in the fourth phase when a credentialed senior figure engages Cofnas at the level his project requires and lands punches. Cofnas has the better case on the 2007 text. Turkheimer has the better case on what good science looks like.
A few patterns run across all four phases.
The forensic auditor mode is the through line. Take the target’s stated position. Reconstruct what the target actually said. Show the gap between the stated position and the reconstruction. The operation works whether the target is MacDonald (right edge of his coalition), Sowell (left edge of his coalition), Carl (internal vanguard discipline), Singer (philosophical mainstream), or Turkheimer (behavior genetics establishment). The instinct is consistent. What changes is what the auditor is willing to do with the audit’s conclusions in different venues.
The procedural-substantive distinction is the early commitment that gradually erodes. The 2020 free-inquiry paper holds the distinction firmly. The 2024 hereditarian revolution piece treats the procedural and substantive as functionally identical. The 2025 MAGA communism piece treats them as distinct again, but in a different direction. The 2026 Substack notes do not respect the distinction. The trajectory across the four phases is toward less distinction, not more. The procedural commitment that licensed the early work has not been abandoned in writing. It has been swamped by substantive commitments that the procedural framework was supposed to keep separate.
The coalition position migrates across the phases. In phase one, Cofnas writes for the heterodox-hereditarian network adjacent to but distinct from the dissident right. In phase two, he writes for that network plus mainstream philosophy of biology, with each side getting the version of him their coalition rewards. In phase three, the coalition splits. The mainstream philosophy of biology network drifts away. The substack-heterodox-hereditarian-Free-Speech-Union coalition becomes the primary audience. In phase four, the X feed and the Substack notes pull him further toward the dissident-right ecosystem (the Doctor-Baron 17cShyteposter repost, the Bronze Age Pervert reply, the carnivore tribal positioning) while his long-form Substack essays continue to mark distance from that ecosystem (Don’t Scapegoat Women, the explicit groyper rejection). The pull and the push are both real.
The hero system consolidates across the phases at increasing intensity. In phase one the persecution narrative is barely visible. In phase two it surfaces in interviews but stays out of the journal articles. In phase three the Andrew Gold catalogue, the Brenner self-disclosure, the Calmversations Korean deplatforming line, the Stanford talk’s Cambridge anecdote, the Boyce Nietzschean self-identification all build the truth-teller-paying-the-cost narrative. In phase four the notes register sustains it daily. The Flemish state television note. The . The Chris Rufo hate mail note. The Turkheimer reply note. Each is a data point in the cosmic narrative. The hero system requires the persecution. The persecution justifies the hero system. The structure is what your Becker analysis correctly identifies, and it has hardened across the four phases rather than weakened.
The metaethical contradiction runs unresolved through all four phases. Cofnas commits to evolutionary debunking of moral intuition in his philosophy of biology work. He commits to confident moral claims about wokeism, free inquiry, intellectual honesty, and the moral failings of his opponents in his popular and interview work. The two commitments cannot both stand. The framework requires moral anti-realism for the philosophical sophistication move and moral realism for the political project. The contradiction has not been addressed in any of the four phases. It surfaces in every interview where a friendly host or a sharp host (Bock, Boyce, Wax, Green, Razib) creates the conditions for it to surface. Cofnas absorbs the surfacing without resolving it. The Trivers self-deception reading covers what is happening. The contradiction would dissolve the project, so the mind protects the project from the recognition.
The trajectory is therefore not a simple decline from rigorous philosophy to coalition advocacy. It is a more complex pattern in which the rigorous philosophy continues but is no longer the center of gravity, the coalition advocacy expands and hardens, the hero system consolidates, and the procedural-substantive distinction that licensed the early work erodes under sustained pressure from the institutional environment and the audience the project now serves. The journal output has slowed but not stopped. The Substack output has expanded heavily. The notes and X output have expanded more heavily still. The follower count plateaus. The subscriber base grows modestly.
The fourth phase contains a few possible futures. One is mechanism. If something like a reliable causal genetic variant for cognitive performance with known effect across environments arrives in the next decade, Turkheimer’s framework predicts that Cofnas wins on the empirical question. The political consequences of that empirical settlement are not what Cofnas’s framework predicts. They are closer to what Turner’s proceduralism predicts. The institutional response to the new empirical situation will be shaped by coalitions and procedures that exist for reasons unrelated to the empirical content of the new finding. Cofnas would be vindicated on the empirical question and probably still excluded from mainstream institutional positions on coalition grounds. The exclusion would not feel like vindication.
A second possible future is institutional drift. The Peterborough ruling has codified part of the architecture under which Cofnas operates. The protected-belief finding gives him legal cover. The hostile-environment finding gives institutions legal cover to keep him out. Future cases will work out the relationship between the two registers. Cofnas’s career might continue cycling through the appointment-petition-protest-counter-petition-institutional-wobble pattern your Hybrid Vigor analysis identifies, with each cycle adding precedent to the legal architecture. The cycle is not stable. At some point the institutional cost of the cycle becomes high enough that institutions stop appointing him. At that point the Substack platform becomes the primary platform. The audience he can reach there is bounded by the platform’s economics, which the current subscriber numbers suggest cannot support a Cambridge-equivalent income.
A third possible future is coalition realignment. That position is structurally unstable. The heterodox-academic-Free-Speech-Union coalition that has hosted him through phase three depends on the woke left remaining the threatening force. If the woke left weakens and the populist right becomes the dominant cultural pressure, Cofnas’s coalition position collides with the new threat environment. He cannot easily migrate further left because his hereditarianism precludes it. He cannot easily migrate further right because the populist-right coalition treats his secular-naturalist commitments and his Korean wife and his Jewish ancestry and his Cambridge credentials as markers of the elite he claims to oppose. The coalition position holds while both threats remain salient. It does not hold under significant change in either direction.
A fourth possible future is the slow accumulation of work that becomes citable when conditions change. The journal output from phase one and phase two has the property that it could survive the disappearance of his current coalition. The 2017 Lorenz paper, the 2016 mismatch paper, the 2018 power-in-cultural-evolution paper, the 2020 incest-taboo paper, the 2024 no-teleology paper. These are real philosophy of biology contributions that engage the existing literature respectfully. They will be cited in twenty years if the conditions for citing them exist. Most of the popular essays won’t survive. The notes will not.
The most striking thing about the trajectory is the speed of the third and fourth phases. Phase one runs three years with relatively few public events. Phase two runs four years with steady journal output and limited public controversy. Phase three runs about two years and produces more public controversy than the previous decade combined. Phase four is six months old and already contains the Ghent petition, the Peterborough ruling, the Turkheimer reply, the Flemish television interview, and the De Standaard montage. The compression suggests the trajectory is accelerating. The institutional environment around hereditarian work is producing more events per unit of time as the underlying conflict intensifies. Cofnas is one of the points where the conflict becomes legible. Whether the trajectory continues at this rate or stabilizes depends on conditions outside his control.
He started as a careful philosopher of biology with a side commitment to heterodox writing, and has become a coalition intellectual whose forensic competence is real but whose self-presentation now varies sharply across registers in patterns the procedural framework cannot quite contain. The journey is not yet complete. The fourth phase is still being written. What the four phases together show is what happens to a serious thinker who refuses crypsis in an environment where most others practice it, who chooses the high-cost move at every juncture, and who continues producing work at multiple registers as the registers diverge under pressure. The result is the figure he is now. The figure he becomes from here will be shaped by events he does not control as much as by choices he makes.

The Four Registers

The communication layer compounds the pressure. Cofnas runs a Substack and an active X account alongside his journal publications. The same argument reads differently in different venues. A hedged paper in Philosophical Psychology reaches specialists who treat it as a contribution to ongoing debate. A punchy Substack post on the hereditarian revolution reaches a larger audience that includes journalists, activists, and politically engaged readers. The 2024 post functioned less as a scholarly argument and more as a coalition signal in cultural politics. The escalation from journal to blog changed how the same underlying commitments were received.
His position in the field has an unusual shape. He has high-status placements in respected journals. He has affiliations at elite institutions. At the same time, he has no large empirical research program of his own to buffer him, no senior chair, no tenured base. The result is a high ratio of visibility to protection. Each controversy carries more consequence than it might for a scholar with deeper institutional anchoring. He moves across philosophy, behavioral genetics, and public writing because he is not embedded in any one of them. That mobility is also his exposure.
Cofnas is strongest as a critic and as a procedural advocate for open inquiry. He is less developed as a constructive theorist of what societies should do if the empirical claims he insists must be investigable were to come out one way or another. If, for the sake of argument, certain group differences with partial genetic bases were robustly established, the policy and ethical implications remain underdeveloped in his writing. Critics use that gap to argue he opens doors without taking responsibility for what comes through them. Supporters argue that the procedural and the substantive should be kept distinct, and that demanding a complete downstream theory before allowing inquiry inverts the normal order of scholarly work. Cofnas sides with the supporters on this point. The argument over whether he is right to do so is still live.
Two commitments hold his career together. The first is his refusal to accept the tacit settlements that allow politically charged empirical questions to be managed without open confrontation. He keeps forcing those questions back into the open, in journals and on Substack, with the same procedural argument and the same auditor’s eye for evidence. The second is his philosophical commitment to a naturalism that does not exempt morality and politics from empirical pressure. Most of his disputes follow from these two commitments held together. Take either one away and the conflicts shrink. Hold both, and they keep producing the pattern his career has by now made familiar.
Cofnas’s academic publications operate at the highest specification density. The 2018 MacDonald paper in Human Nature reconstructs citation chains. The 2022 Utilitas paper “The Golden Rule: A Naturalistic Perspective” reads tradition by tradition, citing Navon’s count of eighty equivalent statements and Csikszentmihalyi on Confucian reciprocity. The 2018 Quarterly Review of Biology paper “Power in Cultural Evolution and the Spread of Prosocial Norms” uses Durham and Fracchia-Lewontin to challenge Boyd-Richerson, then uses Boehm’s reverse dominance hierarchy to argue that majoritarian coalitions deliberately shaped moral norms from the early Middle Paleolithic onward. The 2020 Biology & Philosophy paper on Westermarck engages Lieberman and Lobel’s reanalysis and the Shor-Simchai kibbutz data. The 2024 Biology & Philosophy paper “Natural Selection Requires No Teleology in Addition to Heritable Variation in Fitness” adds a fourth condition to Lewontin’s three, distinguishing natural selection from artificial selection, intelligent design, forward-looking orthogenesis, and selection of nonrandom variation. The August 2024 Substack piece on Sowell catches the Jensen misquote, the Ulster geography slip, and the doctored Cicero passage. These pieces do what their genres do. They specify, footnote, hedge, qualify, and engage the relevant literature in the form the literature recognizes.
The publications also tend toward narrower claims than the popular writings carry. The 2020 free-inquiry paper argues that hypotheses should be evaluated by inspectable evidentiary standards. It does not argue that the equality thesis is wrong, that hereditarianism is correct, or that policy follows. The 2024 Murray-Carl Substack essay presses Carl’s concession on moral worth without claiming the threshold view is unsalvageable in principle. The Conly paper accepts Conly’s case for coercive paternalism and adds intellectual capacity as a premise. The papers stake out limited positions on inspectable terrain.
The popular writings shed most of the qualification. “A Guide for the Hereditarian Revolution” reads as flag rather than as argument. The wokism essay collapses Locke, Christianity, WEIRD psychology, and contemporary progressive culture into a single genealogical line. The MAGA-communism essay deploys an inflammatory label across distinct political tendencies.
The popular writings also carry the coalition-signaling load that the publications do not. “Revolution” rallies allies. “MAGA communism” peels smart anti-wokes from the populist working-class right. The wokism essay flatters secular naturalists over Christians-in-denial. The publications have coalition functions, as the MacDonald and Sowell audits show, but the function operates through the forensic content rather than through the framing vocabulary. The popular writings reverse the ratio. Vocabulary does most of the work. Forensic content, where it exists, supports the vocabulary.
The interviews sit lower on the specification axis than even the popular writings. The Andrew Gold interview accepts the host’s framing about Stuyvesant having been “stolen,” which oversimplifies what has actually happened to the SHSAT and selective programs in New York. Cofnas does not correct it. The interview catalogues the persecution narrative (Daily Mail, Telegraph, Nick Ferrari, Korean deplatforming) at length and lets the catalogue do interpretive work. The Calmversations interview defends the default hypothesis against MacDonald in a register that suffices for the coalition audience without refuting MacDonald on his own terms. Italian fascism before 1938 gets cited as if it equates to German Nazism, which it does not. The Brenner interview surfaces material that the heterodox-right interviewers do not press: that evolution does not select for intelligence, that smart people have fewer children, that the long-run trajectory of intelligence might run against the hereditarian-revolution narrative. Brenner is positioned to ask the question. The other interviewers are not.
The interviews show three further patterns that the publications hide and the popular writings only partly reveal. The first is friendly-host accommodation. Cofnas softens edges, accepts host framings, performs solidarity with the host’s worldview rather than reporting accurately. This is the cost of the format. He is in the room as a guest. The second is identity work the publications cannot do. The Brenner interview surfaces the inherited Jewish communal identity Cofnas is partly exiting and partly retaining. The hereditarian-revolution framework is doing identity-management work that does not appear in the procedural language of the academic papers. The third is hero-system consolidation. Each new institutional rejection becomes a data point. The Korean deplatforming as line on the CV. The Daily Mail fabrication as evidence of persecution. The Strangelove gag closing the Brenner episode. The publications cannot perform this consolidation because the genre forbids it. The popular writings can perform some of it through framing. The interviews perform it most directly because the conversational format rewards autobiographical narrative.
The empirical confidence varies across the four registers. The publications hedge more than the popular writings, and the popular writings hedge more than the interviews and the social media posts. The Andrew Gold interview presents the heritability of group differences as roughly settled. The wokism essay treats the equality thesis as the load-bearing premise of contemporary elite legitimacy. The 2020 free-inquiry paper argues only that the question must remain open. The same author makes all three claims in different venues, and the variation tracks audience expectations rather than evolving evidence.
The political content also varies. The publications mostly avoid direct policy claims. The popular writings sketch policy directions through framing rather than through specification. The interviews state policy positions plainly because the host asks and the format rewards direct answers. Cofnas tells Andrew Gold that he wants to highlight the harm done to White people by the equality thesis. He tells the Calmversations hosts that he expects the establishment cannot keep IQ differences out of the mainstream much longer because the zoomer right knows all about it. The publications do not contain these claims because the publications cannot.
One pattern cuts across all four registers (X, Substack, popular writing, and academic writing). The auditing standard Cofnas applies to opponents (MacDonald, Sowell, Singer, Conly, the Westermarckians, the Boehmians, the moral progress theorists) does not get applied with equal rigor to his own coalition’s central claims. In the publications the asymmetry shows up as topic selection. He audits the equality thesis and not the careful-hereditarian default. In the popular writings the asymmetry shows up as framing. The hereditarian revolution is historical inevitability. The opposing position is theological inheritance. In the interviews the asymmetry shows up as concessions slipping through unnoticed. Evolution does not select for intelligence. WORDSUM data cuts against White race-realists. These admissions sit alongside confident hereditarian claims without integration. The same forensic mind that catches the Cicero passage in Sowell does not catch the trajectory implication in his own concession to Brenner.
The cleanest summary is that the publications protect Cofnas’s epistemic credibility, the popular writings build his coalition, and the interviews consolidate the hero system the coalition needs. Each register does what its genre permits. The three together make the project legible in ways no single register could. Reading only the publications produces a careful philosopher of biology with limited claims. Reading only the popular writings produces a polemical hereditarian with an inevitability narrative. Reading only the interviews produces a persecuted truth-teller with autobiographical color. The three together produce the actual figure: a coalition intellectual whose forensic competence is real, whose coalition position is operative, and whose self-presentation varies by audience.
Looking at the Nathan Cofnas X account on April 28, 2026 shows that many of the items are about Jews. The Galloway reply, the “Twilight of the Liberal Jew” repost, the updated list of politically influential American Jews under 42, and the Steve McGuire repost on Ivy League Jewish political shifts. The hereditarian-revolution framework operates partly as coalition migration from inherited Jewish communal identity to a heterodox-academic position that rewards Jewish-credentialed dissent. The X feed shows the identity work happening continuously rather than residually. He is not occasionally returning to Jewish topics. He is tracking Jewish political behavior as a regular practice.
The “Twilight of the Liberal Jew” updates point at the framework hardening rather than evolving. He wrote the essay in 2023. Two years later he is maintaining and revising the list. The original (under 40) named Shapiro, Raichik, Stephen Miller, Klein, and Zuckerberg. The 2025 update (under 42) adds Sam Altman and Bari Weiss. The list mixes right and center-left. The grouping criterion is ethnicity, not politics. This is a MacDonald move performed in different vocabulary. Cofnas’s 2018 audit refused MacDonald’s group-evolutionary-strategy framework. The X list groups influential Jews by ethnicity for political analysis. The two operations sit closer together than the academic distinction suggests, and the X register reveals the proximity that the publications and even the popular writings can hide.
The Galloway reply shows the journal forensic in a different application. The Moroccan-Jew dilemma is a sharp rhetorical move. If Galloway thinks Moroccan Jews don’t belong in Israel and don’t belong in Morocco, where can they go and what can they eat? The argument is structurally clean, of the kind the MacDonald and Sowell audits perform in long form. On X it operates as victory-by-trap rather than as inspectable argument. Same tool, different use.
The repost of Doctor-Baron 17cShyteposter confirms the bifurcation we sketched between registers. The handle is anonymous-dissident-right adjacent. The careful-hereditarian academic image cannot afford direct association with that ecosystem. On X, Cofnas reposts the account approvingly. The publications keep distance the X feed does not. Both registers exist. The X register is closer to the Substack register than to the journal register, and arguably closer than even the popular writings to the dissident-right ecosystem the academic position has to keep at arm’s length.
The Alan Wolan repost (“pulls zero punches”) is straightforward coalition-signaling and self-promotion. The persecuted-truth-teller hero system the Andrew Gold and Brenner interviews consolidate gets daily reinforcement on X through this kind of approving repost.
The X feed does not contain the auditor mode. No MacDonald-style takedowns. No Sowell-style forensic. No Singer-style tradition-by-tradition reading. The auditing operation that defines his publications and animates his Substack does not appear in the X register. X is for confrontational coalition speech, list-making, identity-tracking, and self-promotion. The bifurcation between his three written registers (publications, popular writings, interviews) extends to a fourth register (X) that operates lower on the specification axis than even the interviews. The same author. Four registers. The differences across the registers are larger than they look from inside any one of them.
The pinned post is “Beating Woke with Facts and Logic” from October 9, 2025. The opening line of the linked Substack is the kind of caricature a journal article would not survive: “When you think of a woke leftist, you might picture a blue-haired, septum-pierced she/they waving a ‘Queers for Palestine’ placard and screaming into a megaphone.” The pin commits the visitor’s first impression to that register. A man who wants to be read as the careful philosopher of biology who audits MacDonald and Sowell pins a culture-war broadside as his front face. The choice tells you which audience the X persona is trying to hold.
The August 30, 2025 post on women in HR (“Women with $65k/year jobs in HR think a plumber making $90k/year isn’t their ‘equal’ because he doesn’t have a BA in sociology”) drew 2.8 million views and 38,000 likes. A day earlier he posted that “a large proportion of perceived ‘sexism’ is when women are treated the same as men.” Then in January 2026 he runs a Substack titled “Don’t Scapegoat Women.” The sequence is the pattern we have been tracking. Maximize provocation, harvest engagement, walk it back when the position becomes inconvenient. Coalition position management in real time.
The February 19, 2026 Substack note on Pinker is theoretically loose in a way the journal papers never are. He argues that the Flynn effect is wrong, that for a period the IQ-120-plus segment seized cultural control, and that “the medieval peasants are still there, just as dumb as ever, and the Internet is bringing them back to power.” The biological claim is incoherent on its face. Population mean IQ does not stay frozen across half a millennium while a thin top crust temporarily seizes control. The Flynn effect either captures a real environmental shift in test performance or it does not. The post substitutes contempt for the masses for analysis. The same author who corrects Sowell’s geography lets this through.
The April 15, 2026 Turkheimer reply is the item that strikes me hardest. Eric Turkheimer is the central living figure on behavior genetics and environmental contributions to heritability.
Here is the initial Cofnas post on April 13. This is the most substantive direct engagement Cofnas has had with a senior figure in behavior genetics. Cofnas has the better case on what Turkheimer’s 2007 article said. Turkheimer has the better case on what good science looks like in the absence of mechanism, and on whether his own current position is what Cofnas claims it is. The exchange ends with Turkheimer offering Cofnas a falsifiable target. Find the mechanism. Until then the dispute does not resolve. The Becker frame reads it as two hero systems colliding. Each man is the truth-teller in his own story.
The November 2024 reply to Bronze Age Pervert preserves a remarkable claim. Cofnas wrote: “the world’s richest man and the incoming Vice President of America both probably know the truth. A single tweet by Musk could open the floodgates.” This is magical thinking about secret elite allies. Musk and Vance silently agree but might at any moment publicly confirm. The hereditarian coalition needs the secret-ally story because the public ally roster is thin. Coalitions under pressure build narratives in which powerful figures secretly support them and might at any moment defect to their side. The defection rarely arrives. The narrative does the work either way.
The X follower count sits around 28k. He posts heavily (over 8,000 posts). The audience does not grow. The hereditarian revolution that the pinned essay declares feasible and desirable is not visible in his own metrics. The cycle is repeat customers rather than expansion.
The Galloway exchange and the “Twilight of the Liberal Jew” updates from July 2025 sit alongside an interview self-promotion from March 2026 (“I was interviewed for Flemish state television“). The persecuted-truth-teller hero system the Andrew Gold and Brenner interviews consolidate gets daily reinforcement on X. Each new media touch confirms relevance. The framework requires the touches.
The X mode is provocation, mockery, list-making, and self-promotion. The auditor mode that defines his publications appears nowhere, including the Turkheimer reply, where it might have served him best.

The Substack notes register sits below the long-form essays on the specification axis and below even the X feed in some moods. The casual cruelty surfaces openly. February 19, 2026: “the medieval peasants are still there, just as dumb as ever, and the Internet is bringing them back to power.” July 9, 2025: conservative-run university centers are “nepotistic (corrupt),” focused on “libertarian propaganda, great books, Christianity, or constitutional history” with “very little serious research.” Jordan Peterson is “clown content” who “could do better than arguing that ‘dragons are real.'” Bookstore “banned books” sections get described casually as “gay porn for children.” The auditor of MacDonald’s footnotes is unlikely to have done forensic work on what specific books appear in specific bookstore “banned books” sections in New York or San Francisco. The phrase is a coalition gesture, not an audit. Groypers get dismissed as too dumb to read 480 words. The literacy gibe is doing the same coalition-discipline work as the right-wing-stupidity essay from January 2024. Cofnas treats inability or unwillingness to read 480 words as evidence of cognitive unfitness. The groypers are dismissed not for the substance of their critique but for the ratio of their attention span to his post length. The same author who treats Sowell carefully on sixty books treats his right-wing critics dismissively on 480 words. The asymmetry tracks coalition lines, not specification density. The “American so-called ‘right’ has become a coalition of stupid people from across the political spectrum.”

The contempt for the masses runs through the Substack notes. Cofnas wants a hereditarian revolution that requires elite acceptance. The notes show him insulting the very populations a coalition would need. Smart anti-wokes are his peers. Everyone else, left and right, gets assessed by IQ first and dismissed accordingly.

The keto and carnivore content catches the eye for what it says about tribal positioning. May 17, 2025: “The keto diet had miraculous benefits for me. I’ve eaten only meat, seafood, and non-root vegetables 7 days a week, 365 days a year since 2019.” Six years of strict carnivore practice. The carnivore community on X overlaps with the dissident-right ecosystem. The diet position carries coalition weight beyond the nutritional claim. A philosopher of biology committing to a contested nutritional position with the same certainty he invests in hereditarianism is making a tribal signal. Notable for someone whose academic project rests on calling out tacit-coalition certainties in others.

The MAGA repositioning matters too. April 2025 brought the MAGA Communism essay separating his coalition from Trump’s working-class right. The Substack notes from that period extend the move. “If I’d known that MAGA would become a poverty cult obsessed with bringing Chinese sweatshops to the US, I would have supported DEI Kamala.” Hyperbolic, but it tells you where his coalition position now stands. He is to Trump’s left on economics and culture, to Trump’s right on race science, attached to a credentialed-dissident coalition that wants to be neither the populist right nor the woke left. This position is structurally unstable. It requires the populist right to stay stupid enough to dismiss and the woke left to stay woke enough to fear. Cofnas has to keep both adversaries in their assigned roles for his coalition to retain coherent boundaries.

The persecution narrative consolidates. March 17, 2026: “I was interviewed for Flemish state television.” Same day: “De Standaard made a montage of me being evil.” December 28, 2025: “Random hate mail from Chris Rufo.” Each note is one more data point in the hero-system narrative the Andrew Gold and Brenner interviews already showed forming. The Substack note format is well suited to this work. Short, image-heavy, easy to repost, low friction. The notes are doing the consolidation that earlier required full essays.

Cofnas’s project depends on his being read across all four registers as the same careful auditor. The registers are diverging. The long-form essays still contain the careful argument. Everywhere else the project is hardening into hereditarian advocacy with culture-war flourishes and a steady supply of persecution content. The brand cannot survive indefinitely on the journal articles alone if the notes, X feed, and interviews keep moving in this direction.

‘Science Is Not Always “Self-Correcting”: Fact–Value Conflation and the Study of Intelligence’ (Feb. 1, 2015)

Cofnas’s argument here is the same engine that drives his MacDonald critique, run on a different target. In both cases he isolates a move where smart people abandon the rules of evidence when the conclusion threatens a coalition value, then he names the move and shows how often it happens. Against MacDonald the move is “ignore Jews who lead opposing movements.” Against the intelligence-research establishment the move is “raise the standard of evidence for hypotheses we find dangerous, and lower it for hypotheses we find flattering.” Same shape.
The Dennett comparison is the cleanest section. Diamond’s hypothesis that New Guineans evolved higher intelligence through tribal warfare gets called magnificent. Lynn’s hypothesis that Europeans evolved higher intelligence through cold-climate selection gets ruled out a priori as awful and racist. The two arguments have the same logical structure. Selection pressure acts on a population over evolutionary time and produces measurable cognitive differences. The asymmetry of treatment is not scientific. It is coalitional. Dennett gives the game away further when he writes that he would be tempted to misrepresent and caricature an idea he found dangerous, and that this is a dirty job somebody has got to do. Then he turns around and condemns Marxists for the identical practice.
The Gould section lands hard because the receipts are so clean. Gould accused Morton of unconsciously cooking his skull measurements to confirm racial hierarchy. Michael remeasured the skulls and found Morton was clean. Gould kept the accusation in the revised edition without mentioning Michael. Lewis and colleagues remeasured half the skulls and found Gould had cooked his own reanalysis by excluding lower-capacity non-Caucasian skulls to lift the non-Caucasian average. Holloway’s line, that Gould was a charlatan whose ideological stance was supreme, is the kind of public statement insiders almost never make. The Mismeasure of Man is taught in courses across the country as a warning about ideologically corrupted science. The irony writes itself.
The Kitcher move on epistemic asymmetry is more interesting than Cofnas gives it credit for. Kitcher’s claim that we should require more evidence for hypotheses with high social cost has a defensible structure if you think of belief as action under uncertainty. Doctors apply something like this when they require stronger evidence to license a risky treatment than to license bed rest. The problem is that Kitcher does not apply the rule symmetrically. Diamond’s flattering hypothesis about New Guineans does not get the higher bar. Only the unflattering hypothesis does. Once the asymmetry shows up the rule collapses into “I will reject what offends my coalition.” Cofnas could have made this point more sharply.
Where Cofnas is weak: the closing rhetoric about truth always winning out is too tidy. Some truths have hurt people in the short and medium run, and the question of who pays the cost of premature publication is harder than his Aristotle quote suggests. He also does not engage seriously with the strongest version of the worry, which is not that scientists know the consequences of their work but that some research programs sit close enough to active political projects that the line between honest inquiry and political ammunition gets thin. That worry does not justify suppressing the research. It does justify asking why a given researcher chose this question rather than another, and what coalition rewards he gets for the choice. Cofnas wants to retire that question. The Alliance Theory angle frame says the question never goes away, on either side.
Through my four-question lens, Cofnas in 2015 is a doctoral candidate at Lingnan funded by a Hong Kong research grant, writing in a small philosophy of science journal, citing Sesardic, Gottfredson, Rushton, Jensen, Lynn, Kanazawa, and Woodley. That is a coalition. It is the heterodox-hereditarian network that sits adjacent to the Pinker-Murray-Reich respectable wing but reaches further out to figures the mainstream treats as untouchable. His income and protection at that point came from advisors and funders inside that network. His status came from being the careful, calm one who could say what the others said without sounding angry. That coalition position does not make him wrong. It does explain which asymmetries he sees clearly and which he glides past. He is sharp on the suppression of hereditarian findings. He is silent on the parallel question of whether some hereditarian researchers have soft-pedaled their own evidentiary problems for coalition reasons.
If race differences in intelligence have a partly genetic component, the path to social justice is harder. That is the real driver of the suppression. Not malice. Not stupidity. A coalition committed to a moral project that depends on a particular empirical premise, defending the premise because the project depends on it. The defense looks like science from inside the coalition and looks like motivated reasoning from outside. Cofnas names the structure without quite naming the coalitional logic underneath it.
The paper holds up. It is the kind of careful, dry, footnoted piece that does more damage to its targets than a thousand polemics, because the targets cannot easily dismiss it as a polemic.

‘Scientific literacy, optimism about science and conservatism‘ (Jan. 28, 2016)

The paper does one useful thing well. It splits a category that gets treated as a unit and shows the pieces behave differently. Self-identified conservatives and social conservatives look less scientifically literate and less optimistic about science than progressives. Economic conservatives, the people who think welfare spending is too high or government should not redistribute, score as well or better than economic leftists on the same measures. The Jost-Mooney story that conservatives are uniformly hostile to science cannot survive that finding.

The cleanest result in the table is the redistribution column. People who oppose government redistribution score higher on the test of the experimental method, on the quiz of scientific facts, on understanding of scientific study, and on the benefits-outweigh-costs measures. They are also less likely to say science makes life change too fast or that we trust too much in science. The pattern is consistent. It points at something the political-psychology literature has mostly ignored: economic conservatism in the American sample tracks with cognitive variables that look closer to libertarian or classical-liberal commitments than to the religious-traditionalist cluster the same surveys lump them with.

The astrology item is the small surprise that cuts the other way. Self-identified conservatives, anti-welfare respondents, and anti-redistribution respondents are all less likely to say astrology is scientific. That fits other survey data showing astrology belief tilts left in the United States, probably because the spiritual-but-not-religious demographic skews progressive. It complicates the simple story that progressives are the science-respecting side.

The methodological limits are real and the authors mostly do not flag them. The scientific-literacy battery is short and weighted toward items that correlate with formal education, so any group with more college exposure looks more literate by construction. The optimism-about-science items mix attitudes that come apart in practice: confidence in the scientific community is a different object from belief that benefits outweigh harms is a different object from “we trust too much in science.” Loading those onto one axis hides the structure. A respondent who thinks vaccines work, climate models are sound, and the NIH is captured by pharma is doing something coherent, but this scale codes him as confused.

The bigger gap is the one the paper gestures at in the discussion and then drops. They cite Duarte and Haidt on academic underrepresentation of conservatives and Cofnas’s own paper on suppressed findings, but they do not push on the obvious next question. If “trust the scientific community” is the optimism measure, and the scientific community has spent decades signaling progressive commitments on the topics where progressives are most invested, then progressive trust in that community is partly trust in a coalition ally. Conservative skepticism of that community is partly skepticism of a coalition opponent. The variable might measure coalition position more than orientation toward inquiry. The same survey data probably shows progressives losing trust in scientific institutions when those institutions tell them something they do not want to hear, on group differences or on COVID origins or on puberty blockers, but the 2000-2014 GSS waves predate most of those tests.

Carl was at Nuffield, Cofnas at Cambridge HPS, Woodley at Chemnitz and the Brussels Center Leo Apostel. All three sit in the heterodox-hereditarian network that overlaps with the London Conference on Intelligence circle. The paper itself is careful and the finding is real, but the choice to run the analysis, the choice of journal (Personality and Individual Differences under Rushton’s long influence), and the framing all sit inside a coalition that benefits from showing the standard story about conservatives and science is wrong. That does not weaken the finding. It does explain why this paper exists and why a parallel paper showing economic progressives outperform economic conservatives on some other literacy measure does not get written by the same authors.

The most useful citation in the paper is Malka and colleagues, the cross-national study showing need for security and certainty correlates positively with social conservatism but negatively with right-wing economic attitudes. That is the empirical point the heterodox literature on political psychology keeps having to relearn. The Jost framework treats conservatism as a unitary motivated-cognition cluster. The data keep saying it is at least two clusters that happen to share a label in American politics for contingent coalition reasons. Carl, Cofnas, and Woodley are reading from that page. The contribution is small. The direction is right.

‘A teleofunctional account of evolutionary mismatch’ (May 6, 2016)

This is a different Cofnas. The other two papers are public-facing arguments aimed at intellectual coalitions. This one is a careful piece of philosophy of biology written for a small specialist audience. The voice is patient, the moves are technical, and the politics drops almost entirely out of view.
The core move is sharp. Lloyd, Wilson, and Sober defined evolutionary mismatch as harmful deviation from the ancestral environment. Cofnas points out that this builds the conclusion into the definition. If you want to study whether deviations from ancestral conditions help or hurt fitness or welfare, you cannot start by defining mismatch as the harmful ones. You smuggle the answer in. So he replaces their definition with a value-neutral one drawn from Neander and Millikan: mismatch is any environmental change that prevents a biological trait from producing the effect it was selected for. Whether the change is good, bad, or mixed becomes an empirical question. That is a genuine philosophical contribution. It makes the concept usable.
The Millikanian apparatus is the engine that lets him cut the territory into four types. Direct proper functions can fail to develop (thalidomide, reading-induced myopia). Direct proper functions can develop fully but fail to operate (peppered moths on soot-darkened bark). Relational proper functions can misrepresent the environment (jackdaws raised by humans courting humans, ducklings imprinting on the wrong species, suburban lawns as supernormal savannah cues). Relational proper functions can use representations correctly but produce responses that no longer reach the invariant goal (apartment cats whose hunting chain comes apart). The taxonomy is clean. Each cell has a working example. The teddy-bear and pornography examples are familiar from Tinbergen and Eibl-Eibesfeldt, but the Millikan framing tightens what they were doing.
The intelligence application is where Cofnas’s coalition reappears, quietly. He uses the new taxonomy to defend Kanazawa, Chiappe, and MacDonald against Penke and colleagues. Critics had said the claim that general intelligence evolved as an adaptation for evolutionary novelty is logically incoherent. Adaptations evolve in response to recurrent features of the environment, so an adaptation for nonrecurrent features is a contradiction. Cofnas answers that novelty can recur. If the environment keeps changing in ways that disrupt existing proper functions, then “ability to handle disruption” becomes a recurrent selection pressure even though the specific disruptions never repeat. That move works. It is the same point Sterelny made about hominin cognitive evolution in The Evolved Apprentice, dressed in Millikan’s vocabulary.
Two weaknesses. First, the framework inherits all the standard problems of selected-effect functionalism. Defining a trait’s proper function as the effect for which it was selected requires you to identify what was selected for. For most complex traits in most species, you cannot identify the selection regime with any confidence. The debates about whether depression is an adaptation, or whether the brain has a domain-general reasoning module, are not debates the Millikan framework can settle. They are debates the framework presupposes settled. So the taxonomy is sharper than the underlying biology will usually support.
Second, the Kanazawa rescue is partial. Cofnas saves the logical coherence of the recurrent-novelty claim. He does not address the harder objection, which is that Kanazawa’s specific applications, especially the claim that liberalism and vegetarianism are evolutionarily novel and therefore correlate with general intelligence, depend on contestable judgments about which features count as novel. Cofnas notes the problem in the Dutton citation but glides past it. The taxonomy gives you vocabulary for the dispute. It does not adjudicate it. Whether contemporary political ideology fits “Misrepresentation-inducing mismatch in the social domain” is exactly the kind of question where the framework is too coarse to compel agreement. Critics will say liberalism is just the cooperative ideology of small-group hunter-gatherers extended; Kanazawa says it is novel; the framework says both have to be analyzed in terms of which proper functions are or are not performed, and now you are arguing about that instead.
The acknowledgments list Sesardic, Millikan, Sober, Zhang. Two reviewers from Biology and Philosophy. This is mainstream philosophy of biology, not the heterodox-hereditarian network. The Sesardic thanks are interesting. Sesardic is a serious philosopher of science who happens to be the most careful defender of behavior-genetic and IQ research against ideological attack, the figure Cofnas leans on heaviest in the 2015 fact-value paper. Their long collaboration sits behind both essays. Cofnas is doing real philosophy in respectable venues while staying loyal to the heterodox network that nurtured him intellectually.
This paper shows the man at his most defensible. He is using technical apparatus he understands to defend a research program he believes in. The coalition payoff is real but understated: a Kanazawa rescue, a Sterelny endorsement, a small advance in evolutionary psychology’s conceptual toolkit. The work could survive the disappearance of his coalition. The 2015 fact-value paper and the 2016 Carl-Cofnas-Woodley paper could not.

‘Innateness as genetic adaptation: Lorenz redivivus (and revised)’ (Jun. 15, 2017)

This is a more ambitious paper than the mismatch piece. The mismatch piece took a working concept and gave it a cleaner definition. This paper takes a concept many philosophers of biology think should be retired and tries to rebuild it on Lorenz’s foundations after Lehrman’s demolition. The bar is higher and Cofnas mostly clears it.
The core move is the right one. Lehrman’s 1953 critique killed the naive Lorenz of the 1930s by pointing out that no trait develops independently of environment, and the deprivation experiment cannot establish what its defenders thought it established. Lorenz’s 1965 reply, which most of the literature ignores, conceded the point and replaced the bad definition with a better one. Innateness is not about traits that develop without environmental input. It is about which source supplies the adaptive information the trait embodies. Two sources, Lorenz said: phylogenetic experience stored in the genome, and individual experience acquired in the lifetime. Cofnas adds the obvious third source Lorenz missed, cultural tradition, and then asks whether the genome-versus-not distinction still picks out something useful once you grant that human adaptations require cultural scaffolding.
The Shea apparatus does the heavy lifting. If you accept Shea’s account of the genome as an inheritance system whose meta-function is to preserve genes correlated with adaptive phenotypes, then genes carry pushmi-pullyu representations: directive content (produce this phenotype) and indicative content (the environment is such that this phenotype fits). The polar bear example makes the abstraction concrete. Gene-W, once it goes to fixation, has the indicative content “the environment is white.” That is what it means for the genome to store information about the environment. Cofnas extends the same framework to cultural variants. If a cultural belief proliferates by natural selection because it tracks fitness-relevant features of the environment, the belief carries indicative content the same way a gene does. The proposition “God hates birth control” is empirically false but, if it spreads because it raises fertility, it functions as a true signal of an environment that supports many children. The signal-truth and proposition-truth come apart. That is a clean philosophical point and a useful one, since it dissolves the apparent paradox in Pinker’s observation that adaptive beliefs can be systematically false.
The overimitation discussion is the best worked example in the paper. Heyes argues overimitation is culturally acquired through mimicry training in infancy. Henrich and others argue it is a genetic adaptation. Cofnas shows the framework handles either answer cleanly. If genetic, the genome stores indicative content about culture itself, namely that all elements of certain models’ goal-directed behavior are worth copying even when their causal relevance is opaque. If cultural, the same content is stored in the cultural tradition rather than the genome. The trait is innate in the first case and not in the second. The empirical question of which obtains is open, but the conceptual distinction is clean. That is the right shape of result for a philosophical paper. It does not adjudicate the empirical question. It shows the empirical question is well-formed.
Where the paper is weakest. The genetic-disease argument runs on a single search of the Journal of Medical Genetics archives and treats the pattern as evidence about how scientists use the word. That is thin. A search of one journal across one specialty does not establish much about scientific usage in general, and the philosophical literature treating genetic disease as a paradigm case of innateness has its own reasons for that treatment that Cofnas does not engage. He gestures at the discrepancy and moves on. The Mameli and Bateson cluster account, which says innateness corresponds to several only loosely correlated properties (i-properties), gets a faster dismissal than it deserves. Their argument is that even if innateness has a meaning in some uses, the term encourages illegitimate inferences from one i-property to another, which is a charge about the word’s pragmatic effect on scientific reasoning rather than about its semantic content. Cofnas’s reply is that scientists use the word to mark genetic adaptation specifically, so the bad inferences they worry about do not actually happen. That reply needs more empirical support than the Journal of Medical Genetics search provides.
The epigenetic section is short for a reason. Haig’s point that the capacity for epigenetic switching is itself a genetic adaptation does most of the work, and Cofnas leans on it without overstating. The honest concession is that if Jablonka and Lamb turn out to be right about high-fidelity epigenetic inheritance accumulating adaptive information across generations, the account needs revision. He says so. That is the right tone.
The acknowledgments tell a story. Patrick Bateson, Cecilia Heyes, Tim Lewens, David Papineau, Neven Sesardic, Kim Sterelny, two BPhil reviewers. Bateson is one of the philosophers Cofnas is arguing against, and getting his comments on a draft is the kind of thing that happens when you are doing real philosophy in good faith. Heyes is the source of the cultural-overimitation hypothesis Cofnas treats as live. Lewens runs Cambridge HPS and is the leading philosopher of cultural evolution in the UK. Papineau is the senior figure in teleosemantics. Sterelny is the figure whose Evolved Apprentice provides the apprentice-learning framework. This is the cream of the discipline, and the paper reads like it was sharpened by their criticism. Sesardic is the only name from the heterodox network Cofnas is loyal to elsewhere. The center of gravity has shifted toward mainstream philosophy of biology.
The Cofnas of 2017 sits at Darwin College Cambridge. His income, status, and protection now come from the Cambridge HPS world: Lewens, Papineau, the Biology and Philosophy reviewers, the Springer publishing apparatus. The coalition rewards careful Millikan-style work that engages the existing literature respectfully. The paper delivers exactly that. He is no longer just the Sesardic protégé who took down Kevin MacDonald. He is becoming a recognizable figure in mainstream philosophy of mind and biology. The same instrument that produced the heterodox-coded fact-value paper in 2015 is here producing a paper any philosophy of biology department would be happy to claim. That is what intellectual mobility looks like in academic philosophy. The four pieces together let you see a man building two careers at once: the public-facing heterodox writer and the credentialed analytic philosopher of biology. The credentialed track requires constant deposits of work like this paper. Cofnas is making the deposits.
The most useful single contribution is the analysis of how culturally transmitted variants can carry indicative content in the same way genes do, and the resulting clarification that “innate” should pick out the genome-as-source case specifically rather than phylogenetic adaptation broadly. That move both saves Lorenz from his own oversight and gives scientists like Spelke, Haidt, and Wertz a defense against developmental-systems theorists who would have them retire the word. Whether the move ultimately persuades philosophers committed to abandoning innateness is a separate question. The argument is honest, the apparatus is appropriate, and the paper deserves to be in the literature.

‘Religious authority and the transmission of abstract god concepts’ Sep. 15, 2017

Cofnas identifies a real hole in the Standard Model. The SM treats religion as a contest among MCI concepts for memorability and transmission. He correctly notes this misses what laypeople do most of the time: defer to religious experts whose doctrines they cannot articulate or even understand. The Catholic in the pew does not believe in an MCI agent who answers prayers sequentially. He believes his priest knows more than he does about what God is, and accepts that authority structure.
The critique of the Barrett and Keil evidence base lands. A handful of studies on prayer-vignettes cannot carry the weight the SM places on them. And Cofnas’s point about transmission is sharp: what passes between people is the verbal explicit content, not the implicit representation. If the explicit content is highly counterintuitive (omnipotence, omniscience, immateriality), then the SM is at best a theory of how that content gets distorted in private cognition, not a theory of cultural transmission.
But the positive account is weaker than the critique. Cofnas wants to say people accept doctrinal religions through “the plausibility of what they can understand and the intellectual credibility of the experts.” This relocates the question rather than answering it. What confers intellectual credibility? Why does Pat Robertson get to keep his coalition if he switches to Catholicism but lose it all if he switches to Zeus?
His own Pat Robertson example points toward an answer he does not develop. The reason Robertson cannot pivot to Mickey Mouse, even with maximum CREDs, has little to do with logic and much to do with coalition structure. His audience selected him as a model because he speaks for a coalition with stakes in particular doctrines. A switch to Mickey Mouse is not a change of conclusion. It is an exit from the coalition. The doctrines mark the boundary. CREDs and prestige operate inside that boundary, not before it.
This is where Alliance Theory angle does work that Cofnas’s “logical development” framing cannot. Doctrines proliferate because they coordinate coalitions, and theologians develop them under pressure to maintain coherence as the coalition expands across populations. The free will and omnipotence paradox is a case in point. Slone argues theologians independently arrived at the same three solutions because the logic of the paradox allows only those solutions. Maybe. But the paradox only arises if you have already committed to two coalition-sustaining premises: an all-powerful God responsible for everything (which underwrites priestly authority and ritual significance) and human responsibility that carries moral weight (which underwrites moral teaching and behavioral compliance). The parallels across Judaism, Christianity, Islam, Hindu traditions, and Buddhism reflect parallel coalition pressures, not free-floating logic. Drop either premise and you lose the apparatus that holds the coalition together.
The Judaism and Hinduism convergence on a self-existent creator who manifests in but stands outside time is striking. Cofnas reads this as theologians reasoning their way to similar conclusions from similar starting points. Becker’s The Denial of Death points another way. Both traditions need a hero system that can absorb death, and a fully transcendent creator does that work in ways an MCI agent cannot. The technical philosophical apparatus is what specialists produce when they must defend the transcendent claim against scrutiny. The lay believer does not need the apparatus. He needs to know that someone trustworthy has it covered.
The piece also undersells Whitehouse. The doctrinal-mode account already explains much of what Cofnas claims the SM misses: the routinized rituals, the priestly interpretive monopoly, the logical coherence pressure on theology, the spread across anonymous populations. Cofnas cites Whitehouse but does not credit how much of his own positive picture is borrowed. The contribution that feels new is the explicit framing of expert deference as a missing variable in CSR. That part deserves the attention it gets.
Cofnas treats the analogy between religious and scientific authority as roughly symmetric: laypeople defer in both cases because they recognize superior expertise. Stephen Turner’s work on expertise and democracy complicates this. Religious authority and scientific authority are both achieved through coalition processes, not through some pure recognition of competence. Turner’s point about epistemic coercion is that what looks like deference to expertise is often deference to whoever has the institutional standing to define what counts as expertise. That cuts against Cofnas’s clean analogy and points back to the coalition account.

‘Does Activism in Social Science Explain Conservatives’ Distrust of Scientists?’ (March 2018)

The paper makes one sharp move and several softer ones. The sharp move separates trust in scientists from trust in science. Gauchat measures the first and treats it as the second. Cofnas, Carl, and Woodley catch him on it. The McCright finding closes the door on a lazy reading. Conservatives trust food and materials scientists more than liberals do. If conservatives distrusted science as such, they would not trust production scientists more.
Putnam supplies their best evidence. He admits he held findings until he could append a liberal-friendly forecast. He says publishing without that move would have been irresponsible. He then accuses scholars who cite his short-term findings without his speculative long-term forecast of selective citation. That catches him cleanly. The amicus brief from the ASA and APSA dismissing Putnam’s data on procedural grounds while building its own case on a “well-established body of literature” catches them too.
The 1000-studies myth on media violence holds up. Major medical and psychological associations testified to Congress that thousands of studies supported a causal link they had not reviewed. Freedman counted around two hundred. The associations put their authority behind a number wrong by an order of magnitude.
Where the paper softens. It documents a pattern of activism but does not show that the pattern explains the time series. The GSS trend sits on one axis and a list of activism cases on the other. The authors never connect the two with evidence that conservatives know about these cases and update on them. The argument is plausible. It is a hypothesis dressed as a finding.
The paper also fails to check itself. Cofnas, Carl, and Woodley sit in a particular coalition, the heterodox-hereditarian wing that includes Sesardic, Gottfredson, Jensen, and Rushton. The papers they cite as victims of censorship are mostly papers that coalition supports. The Sternberg and Gardner quotes they wave concern race and IQ, the coalition’s flagship file. Run my four questions on the authors. Who do they rely on for status? Heterodox networks, IQ research outlets, Quillette-adjacent venues. Who do they need to retain? The same readers. What signals membership? Treating Summers as martyr, treating Stapel and LaCour as representative, treating stereotype threat as the paradigm failed liberal effect. What would they lose by reversing position? Their entire intellectual home.
That does not make the case wrong. Putnam did what he did. But the paper performs the same selective citation it charges the field with. It picks cases that support its conclusion. It does not look hard for production-scientist scandals or conservative-friendly findings that survived peer review without trouble.
The Turner connection is what the paper misses. Turner’s work on expertise and democratic legitimacy is the frame this argument needs. Scientists claim authority to settle public questions. That authority rests on a perception of restraint. Activism depletes the credit. The loss of credit shows up in survey data. The paper gestures at this without naming it. Weber’s “Science as a Vocation” appears once, late, as a closing flourish.
The trust-in-scientists frame holds. The Putnam evidence holds. The causal story across forty years of GSS data is asserted rather than tested. And the authors’ coalition position is invisible to them, which is the same charge they level at the field.

‘Is vegetarianism healthy for children?’ (Feb. 23, 2018)

The strongest material is the Kenyan study and the creatine work. The Neumann trial gave roughly equicaloric supplements of meat, milk, or oil to children whose baseline diet was nearly vegetarian. The meat group outperformed the milk group on fluid intelligence by about 0.65 SD, on arithmetic, on muscle gain, and on physical and social activity. This is the only controlled comparison of meat versus milk supplementation in children, and the AND (Academy of Nutrition and Dietetics) ignores it. That omission is hard to defend on scientific grounds. The creatine evidence from Rae et al. and Rawson et al. fits the same pattern. Vegetarians improved roughly 1 SD on fluid intelligence and working memory tasks with creatine supplementation, while meat-eaters showed no benefit. The natural reading is that the vegetarian baseline reflects deficiency, not a ceiling effect.
The B12 numbers also land. Herrmann’s findings, where vegans averaged 148 pmol/L and even supplementing vegans clustered in the marginal range, sit poorly with the AND’s confident statement. The childhood consequences of B12 deficiency can be permanent. If the AND is going to make a blanket claim that vegan diets work for “all stages of the life cycle,” the burden of evidence on this point is high, and the available data does not meet it.
The hypospadias finding from North and Golding is striking, a 3.5x relative risk linked to phytoestrogen exposure in pregnant vegetarian mothers. Whether that replicates is another question, but the AND ignoring it while recommending soy as the primary meat substitute for pregnant women is the kind of thing Cofnas is right to flag.
Now the weaker parts. The Hudson and Buckley sex ratio finding is one paper at one British hospital with a small vegetarian subsample. Cofnas treats it more confidently than the evidence warrants. Sex ratio at birth is noisy data and the malnutrition-stress-male-loss chain has many competing explanations. He should have cited it as suggestive at most.
The Gale et al. detour about vegetarian IQ and redistributionist politics, with the swerve into Kanazawa and Carl on intelligence and political views, is the weakest passage in the paper. It has nothing to do with whether vegetarianism is healthy for children. It reads like an aside the author could not resist. It opens flanks that distract from the core argument and invites readers to dismiss the rest as ideologically motivated. Kanazawa’s work on intelligence and ideology has serious methodological problems and citing it weakens rather than strengthens his case. A careful editor would have cut that paragraph entirely.
The acne section is reasonable as far as it goes but the link to the vegetarian question is thin. The argument is: dairy causes acne, vegetarians who substitute dairy for meat consume more dairy, therefore vegetarian children have more acne. The middle premise is assumed rather than demonstrated. Many lacto-ovo vegetarians do not increase dairy intake.
The taurine and DHA sections are honest about uncertainty. He repeatedly says the clinical relevance is unknown, which is the correct posture. The point that the AND itself uses the word “unknown” while still issuing a confident recommendation is a fair catch.
The deeper structural critique is the one worth lifting out. The AND is a professional body whose members have strong commitments to dietary recommendations that align with environmental and ethical concerns prominent in their professional coalition. The position statement reads like advocacy dressed as science. That is a coalition-maintenance document, not a literature review. The four diagnostic questions apply. Who do the AND authors rely on for status and protection? The professional and ideological networks that endorse plant-based eating. What signals coalition membership? Position statements that affirm vegetarianism’s safety across the life cycle. What would they lose by qualifying the claim? Standing within those networks and the ability to serve as expert witnesses for plant-based advocacy. The science-policy gap Cofnas documents is what coalition pressure on professional bodies looks like in practice.
His paper has had 88,000 views. That is unusual for a nutrition review. The reason is that he said something the AND coalition has institutional reasons not to say, and parents searching for honest information about feeding their children find very little of it. The piece is useful for that reason even with its flaws.
The political detour should come out and the sex ratio claim should be softened. Otherwise the argument holds.

‘Larregue’s Critique of Cofnas et al. (Mar. 12, 2018): A Rejoinder’

The rejoinder lands several clean hits and exposes Larregue as the weaker debater. Cofnas, Carl, and Woodley catch him misquoting them on “affirmative action,” a phrase that appears nowhere in their paper. They catch him on the Pioneer Fund claim about Herrnstein and Murray, who never received that funding. They catch him calling Heterodox Academy a conservative outfit when its own member survey shows roughly equal numbers of left and right identifiers. Each of these is a basic factual error in a published critique, and each reveals the kind of carelessness that comes from coalition-driven reading rather than close reading.
The strongest move in the rejoinder is the burden-shift on cherry-picking. If Cofnas et al. cherry-picked, Larregue needed to produce counter-examples. He needed cases where the ASA, the APA, or the APSA filed an amicus brief or testified to Congress in favor of a conservative position. He produced none. The challenge sits there in the text and Larregue cannot answer it. That silence does real work for the authors.
The contexts of discovery and justification move also lands. Larregue accuses the authors of conflating motivation with truth, then turns around and tries to discredit them by listing the conservative outlets where Cofnas has published, the podcaster who interviewed Woodley, and the blog where Carl has written. That is the conflation Larregue named as a fallacy three pages earlier. The authors catch this with the right amount of edge, not too much.
Where the rejoinder shows strain. The causal claim still floats. Cofnas concedes activism is “difficult to measure” and that proving the link to GSS trends is “difficult.” His fallback is that the absence of comparable conservative cases makes liberal drift the reasonable explanation. That is a defensible position for an essay. It is not a tested hypothesis. The original paper sold the activism explanation as a finding. The rejoinder quietly downgrades it to a reasonable inference. A reader who liked the first paper has to notice the retreat.
The Inbar and Lammers exchange is murkier than the authors admit. Larregue’s point was that Inbar and Lammers themselves do not draw the conclusion Cofnas et al. draw from their data. That is a fair point about how the evidence is being used, and the rejoinder dodges it by saying Cofnas et al. are making a different argument from the survey data than Inbar and Lammers made. That is true but it does not address whether the inference from willingness-to-discriminate-in-surveys to actual-distortion-of-published-science holds up. Surveys of stated willingness are not measurements of behavior.
The Stefan Molyneux defense is weak. Saying Molyneux also interviewed Turkheimer and Flynn does not establish him as politically neutral. Molyneux ran a YouTube channel with a clear ideological project, and being interviewed there carried signal about which audience a scholar was willing to address. The authors are right that an interview is not proof of bias. They overplay the response by suggesting Molyneux is some kind of neutral interviewer. He was not.
The Hawking correction is interesting. Cofnas is right that Hawking’s specific Venus claim was scientifically indefensible and was criticized by climate scientists. Larregue elides that to portray Cofnas as a climate denier. The rejoinder catches the elision. This is the kind of detail that matters. A serious critic should have read the Weekly Standard piece before citing it. Larregue did not, or did and misrepresented it.
The deeper issue the rejoinder does not face. Larregue’s coalition-membership argument is wrong in form because it commits the genetic fallacy. It is right in substance because it identifies the small heterodox-hereditarian network these three authors inhabit. The Pioneer Fund link the authors object to is a real link in their intellectual ecosystem even if Herrnstein and Murray did not personally take the money. Run my four questions on Cofnas, Carl, and Woodley and the same pattern shows up that showed up in the first paper. They write for the same outlets, cite the same authors, defend the same victims, and treat the same enemies as enemies. The rejoinder is well-argued within its frame. The frame stays invisible to the authors.
Two papers, taken together. The first paper makes a defensible distinction Gauchat missed and supplies real evidence on Putnam, the AAP, and Summers. The rejoinder defends that core while quietly retreating on the causal claim. Larregue gives the authors easy targets by getting facts wrong, and they hit those targets cleanly. The exchange leaves the strong original points intact. The weak ones, scaled-back but unrefuted, sit where they sat.

‘Judaism as a Group Evolutionary Strategy: A Critical Analysis of Kevin MacDonald’s Theory’ (March 10, 2018)

The paper’s strongest move is the unfalsifiability charge. MacDonald reads any pattern as confirmation. Jewish leadership of a movement shows ethnic activism. Gentile leadership shows that gentiles got recruited as front men. Jewish opposition to anti-Semitic movements shows extreme ethnocentrism. Jewish opposition to Israel or to Jewish interests, like Chomsky or Soros, gets ignored or footnoted away. Once a theory absorbs every possible observation, it stops doing work.
The Marcuse passage Cofnas quotes does serious damage. Marcuse argued for Arab return to Israel and against a permanent Jewish majority. MacDonald casts Marcuse as a Frankfurt School exemplar of Jewish hypocrisy, multiculturalism abroad and ethnic purity at home. The record shows the opposite. Same for Fromm. The hypocrisy charge requires identifying individuals who hold inconsistent positions, not aggregating across different Jews who disagree with each other and calling the resulting contradiction proof of a double standard.
The Sanford misrepresentation is clean. Sanford was a gentile. The passage MacDonald quotes about conformity has Sanford distinguishing nominal from genuine Christianity, with the genuine version scoring low on ethnocentrism. Sanford was praising Christian humanism, not denigrating Christianity. MacDonald flips it. That kind of sourcing problem, repeated across the book, kills the project as scholarship.
Cofnas’s default hypothesis handles the data more cleanly. High IQ plus urban concentration predicts overrepresentation in any non-anti-Semitic movement. That covers chess champions, Nobel laureates, neoconservatives, the National Association of Scholars, FIRE, paleo-conservatives like Paul Gottfried, and the occasional Jewish presence at American Renaissance. MacDonald’s theory requires a directional skew toward leftism that the data does not bear out once you count carefully. The same population shows up wherever prestigious cognitive coalitions form.
What Cofnas does not quite confront is that the underlying question can be reformulated in coalition terms without MacDonald’s essentialism. Different Jewish sub-coalitions back different movements. Reform Jews in midcentury New York worked one set of alliances. Orthodox Jews in Brooklyn work different ones. Israeli right-nationalists work yet another. Treating the Jews as a unified actor is the Turner essentialism error in pure form. Drop it and the residual question becomes which sub-coalitions cluster where, and why. That question is tractable. MacDonald’s is not.
There is also the question of why MacDonald’s project keeps regenerating among readers despite the scholarly demolition. The alt-right does not read MacDonald for testable propositions. It reads him the way any coalition reads its founding texts, for moral vocabulary that justifies the alliance. The book provides a frame in which any Jewish behavior confirms prior commitments. Falsification is beside the point because the function is coalition-maintaining, not predictive. Pinsof’s Alliance Theory predicts the persistence of the book even after the academic demolition. Becker’s hero systems predict the emotional intensity of the readership.

‘What the Alt-Right Gets Wrong About Jews’ (Mar. 15, 2018)

This Quillette piece is the same argument as the academic papers compressed into popular form, with the gloves slightly off. The interesting thing about it is what gets added and what gets dropped.
What gets added is the explicit causal flip. The academic papers gesture at this. The Quillette piece names it. MacDonald says Jewish liberalism causes anti-Semitism. Cofnas and Anomaly say anti-Semitism causes Jewish liberalism. Persecuted minorities gravitate toward political philosophies that emphasize social tolerance and free movement of people because those philosophies protect them. The Holocaust reinforced the lesson. This is a cleaner thesis than the one in the academic papers, which spread the explanation across IQ, urban concentration, geography, and persecution. Here the persecution mechanism is doing more work, and the piece is stronger for it.
The Mark Twain framing at the top is rhetorically smart. Twain asks why Jews are heard of out of proportion to their bulk. The conspiratorial answer fell out of fashion after 1945 and came back through MacDonald. Naming this lineage matters because it situates MacDonald not as a brave heterodox scientist but as the latest retail vendor of a very old product.
The Edmund Burke quotation is the best move in the piece. Burke describes radical clubs deforming public measures, academies as seminaries for these clubs, daring and violent counsel taken as the mark of superior genius, tenderness to individuals treated as treason to the public. This is 1790. There are no Jews in the picture. The thing MacDonald describes as the distinctive Jewish intellectual style is just the radical intellectual style, and the radical intellectual style predates Jewish entry into European intellectual life by centuries. The French Revolution itself, the Sartre-Beauvoir-Camus circle, Foucault. Cofnas and Anomaly compress what took the academic papers many pages to argue into three or four paragraphs and the compression strengthens it.
The libertarian list is the move that does the most damage in the smallest space. Friedman, Mises, Nozick, Rand, Rothbard, Kirzner. The anti-communist list. Stephen Miller as Trump’s senior policy advisor and most influential anti-immigration activist. If Jews are pursuing a unified group evolutionary strategy, why are Jews running both sides of every fight? The default hypothesis answers this in one sentence. MacDonald has to answer it with thirty pages of ad hoc reasoning about which Jews are real Jews and which Jews are crypto-activists and which gentiles are puppets and which gentile-led movements somehow trace back to Jewish influence anyway. By the time he gets done, the theory is wallpaper.
The Popper-Kuhn-Lakatos closing is correct but a little too gentle. They say no theory is strictly falsifiable, you have to use judgment, and judgment says MacDonald’s ad hoc patches have piled up too high. This is right. But the stronger claim, which the academic papers eventually arrive at and this piece declines to make, is that MacDonald’s theory does not have a falsifiability problem in the ordinary sense. It has the structure of a coalition-justifying narrative rather than a predictive one. The closing line about the alt-right needing an enemy and Jews being a convenient people to play the role gestures at this but doesn’t quite say it. The piece is for Quillette readers and Quillette readers want science-versus-pseudoscience framing, so the authors give them science-versus-pseudoscience framing. The deeper structural point, that MacDonald’s book performs a coalition function and would not perform it if it were falsifiable, gets left at the door.
What gets dropped from the academic version is most of the Sarfatti, Sanford, Graham, and PCIN sourcing material. The popular piece can’t run those receipts in detail. It loses force as a result. A reader of the Quillette piece alone could come away thinking this is a he-said-she-said about competing interpretations. A reader of the academic papers knows MacDonald cited Sanford for the opposite of what Sanford said, cited Graham for the opposite of what Graham said, cited a PCIN page that does not contain what he claimed it contained, then a different page, then a third page, none of which contained it. The popular piece soft-pedals the misconduct.
The Anomaly co-authorship is worth noting. Anomaly is a philosopher of biology who later moved into hereditarian race-and-IQ work and into bioethics work that includes some unusually frank treatments of eugenics. Like Cofnas, he is signaling here that you can reject MacDonald without rejecting the broader hereditarian project. The Cochran-Hardy-Harpending citation does that work. Ashkenazi IQ evolved through medieval selection on white-collar professions. This is the standard hereditarian story. It is presented as established. The piece thereby positions itself as race-realist about IQ but anti-MacDonald about Jewish conspiracy. That positioning is the same one Cofnas takes in the 2021 paper. The two-front war framing.
A small thing the piece gets exactly right that the academic papers also got right but buried. The Holocaust did not teach Jews that liberal cosmopolitanism is metaphysically true. It taught them that nationalist movements are dangerous to them. This is a coalition-formation lesson, not a philosophical conversion. Jewish liberalism is not Jews discovering universal truth. It is Jews learning what kinds of political arrangements protect them and aligning with those arrangements. The same Jews under different conditions, Israeli Jews under existential threat from neighbors, vote nationalist. American Orthodox Jews under threat from secular progressives now vote Republican. Soviet refusenik Jews became American conservatives. The political behavior of Jewish sub-coalitions tracks the threat environment. There is no unified Jewish strategy. There are particular Jewish coalitions reading particular threat environments and aligning with whoever offers protection. This is the Pinsof move and it is the right one. The Quillette piece almost makes it but stays at the level of “persecuted minorities tend toward liberal cosmopolitanism” rather than developing the coalition-conditional version.
The piece is good for its venue. It will reach readers who would never read Human Nature or Philosophia. It does not pretend to neutrality the way the academic papers do, which is a virtue. The thing it does not do, and which neither the academic papers nor anyone else has yet done well, is explain why MacDonald’s book keeps regenerating among readers who have access to the demolition.

MacDonald’s work on Jews is not failing at what its readers want it to do. It is succeeding.

A scholarly book and a coalition document are scored differently. A scholarly book is scored on whether its claims hold up. A coalition document is scored on whether it provides moral vocabulary, identifies friends and enemies, and explains the world in a way that justifies the alliance. By the first standard MacDonald failed. By the second standard he is doing exactly what his readers need him to do, and Cofnas’s demolition is irrelevant to that function.

The alt-right has a problem its founding documents have to solve. The problem is that the explicit white nationalism it wants is morally illegible to most Whites. You cannot recruit a mass white movement on the basis of “we want power for our race” because most Whites in the post-1945 West find that framing repulsive. You need a story that converts the desire for white power into the desire for white self-defense. You need an enemy whose existence makes white solidarity defensive rather than aggressive. You need that enemy to be small enough to plausibly defeat, large enough to plausibly threaten, prestigious enough to explain why Whites do not currently rule, and cohesive enough to count as one actor.

Jews fit every requirement. MacDonald’s book is the document that makes the fit explicit. Without something like MacDonald’s book the alt-right has to admit it is asking Whites to dominate other people. With MacDonald’s book it can say it is asking Whites to defend themselves against an actual unified hostile coalition. The moral vocabulary changes from offense to defense. That change is not optional. It is structurally required for the movement to recruit.

This is why the demolition does not bite. Cofnas shows that Marcuse advocated Arab return to Israel, that Sarfatti built Italian fascism, that Reform Jews lobby for racial diversification of the Jewish community, that Sweden became extreme without Jews, that Hugh Davis Graham said the opposite of what MacDonald cited him as saying. None of this matters to the function. The reader does not need the book to be true. The reader needs the book to exist. The book’s existence is what licenses the moral framing. As long as a thick, footnoted, academic-credentialed text says the thing, the coalition can point to it and say see, we are not just bigots, we are responding to documented patterns. The footnotes are decoration. The function is laundering.

This is why MacDonald’s response to Cofnas was so thin. He is not really arguing scholarship. He cannot afford to lose the credentialed-text status of the book, but he also does not need to win the scholarly debate to keep the coalition function. The retraction of his Philosophia reply actually helped him on coalition terms. Suppression confers martyrdom. The coalition document becomes more potent when persecuted, not less. Nietzsche’s line about the world-historical stupidity of all persecutors is exactly right here, and exactly inverted from what Cofnas hopes it means. Cofnas thought he was warning his side not to suppress MacDonald because suppression looks bad. The deeper thing is that suppression is irrelevant to truth and converts directly into coalition fuel.

The 1798 to 1945 European tradition of anti-Semitism worked the same way. Lueger said anti-Semitism is a sport for the common people, useful for getting ahead in politics, and once you are up there you do not need it any more. He was telling on the function. The Protocols of the Elders of Zion is the obvious case. The Protocols are a forgery. This was established almost immediately. The book sold tens of millions of copies anyway. Henry Ford printed half a million in America. Nazi schools assigned it. The forgery question never mattered because the function was never to be true. The function was to provide a comprehensive narrative explaining every grievance through one cause, and every grievance can be explained through any cause that is salient enough and unified enough. Once you have the narrative, refutation just looks like the enemy defending itself.

MacDonald is the academic-respectable version of the same product. A reader who would feel ashamed citing the Protocols will cite MacDonald without shame because MacDonald has a CV, references, footnotes, university affiliation, peer review history. The coalition gets the same narrative content with cleaner sourcing optics. This is the upgrade. The narrative was always going to exist in some form. MacDonald just produced the version that cleared a higher reading-class bar. Spencer’s line about MacDonald being the most essential man in his movement in terms of thought leadership is correct on coalition grounds. MacDonald did not invent the thoughts. He gave them academic packaging.

There is a second function that operates underneath the first. The book provides emotional integration for a particular kind of reader. The reader is usually an intelligent White man who has noticed disparities between his abilities and his social position, between Jewish overrepresentation in elite positions and his own coalition’s political weakness, between the moral rules applied to his group and the moral rules applied to other groups. He has noticed correctly. The disparities are real. He needs an explanation. The mainstream offers him two: Whites are over-represented through historical privilege and need to make room, or Whites are equally talented but face no discrimination and just need to work harder. Neither explanation accounts for what he actually sees. MacDonald offers a third: there is an organized opposing coalition whose interests run contrary to yours and whose success is causing your difficulty. This explanation has the virtue of being structurally similar to the truth. There are coalitions. They do compete. Their interests do diverge. Your difficulty is not entirely your fault. Where MacDonald goes wrong is in essentializing one of those coalitions and unifying it across time and geography in ways the evidence will not support. But the wrongness is at the level of detail. The basic structure, that you are losing because an organized opposing coalition is winning, is closer to right than either of the mainstream alternatives. This is why the book lands.

A reader who picks up Cofnas’s demolition learns that MacDonald got the details wrong. He does not learn that the mainstream alternatives got the structure right. The mainstream alternatives still do not account for what he sees. So the demolition leaves him with the original problem and one fewer answer. He goes back to MacDonald because MacDonald is still the only document on offer that takes the structure of coalition competition seriously. Cofnas’s default hypothesis, high IQ plus urban concentration plus persecution producing leftward tilt, is technically a coalition-aware analysis but it is presented in such a way that it dissolves the coalition rather than naming it. The reader does not feel his situation explained. He feels it explained away.

There is a coalition, it has interests, it competes with other coalitions, its members are not all the same, the coalition’s boundaries shift over time, particular sub-coalitions ally with particular outside sub-coalitions for particular reasons. This is the analysis MacDonald should have done and could not do because his readers do not want sub-coalitions and shifting alliances. They want one enemy. The mainstream race narrative makes the same essentializing move from the other direction. Whites are one enemy. Both narratives are essentialist for the same reason. Coalition-justifying narratives have to be essentialist or they cannot do their work.

MacDonald’s book regenerates because the demolition addresses scholarship and the readership wants something else. The readership wants moral cover for white solidarity, emotional integration of real grievances, and a comprehensive narrative that explains observed disparities through coalition competition. The book provides all three. The footnote errors do not affect any of those three functions. The book will keep regenerating until someone produces a comparable document that takes coalition competition seriously without essentializing the coalitions, and that someone has to be willing to lose the readership that wants essentialism. Almost no one in this discourse is willing to lose that readership. Cofnas tried for two papers and then drifted toward a different essentializing coalition. MacDonald never tried. The mainstream race narrative does not try because it benefits from essentializing in the opposite direction. There is no single book yet that does for coalition analysis what MacDonald did for essentialist anti-Semitism. Until there is, MacDonald keeps the field by default.

The wars are real. So, possibly, is what the combatants are fighting about. But the combatants are not who MacDonald says they are, and not who his critics say they are either. The combatants are particular sub-coalitions in particular cities at particular moments making particular alliances. Naming them at that resolution is the work. Whoever does that work writes the book that finally displaces MacDonald.

Five distinguishable coalitions are advancing anti-Jewish discourse in America right now. They share almost nothing except the rhetorical surface. Treating them as one movement is the first analytical mistake. Each has different interests, different recruits, different funders, different theories of victory.
The first is the dissident right intellectual coalition. MacDonald, the Occidental Observer, Counter-Currents, the Unz Review, Steve Sailer’s circle, the magazines and substacks that orbit them. These men are mostly highly educated, mostly White, mostly downwardly mobile relative to their abilities or their fathers. They are losing status competitions to people they consider less able than themselves and they have noticed that Jews are overrepresented among the winners of those competitions. The coalition’s function is to explain that loss in a way that does not require its members to accept either mainstream story. The interest served is moral. They get to be defenders of their people rather than failed competitors. The coalition has almost no political power and almost no money. Its product is text. Its readership is small but devoted. It produces the intellectual scaffolding the cruder coalitions later strip for parts.
The second is the streamer-podcaster coalition. Nick Fuentes, the Groypers, Sneako, Jake Shields, the post-Trump online right that broke from mainstream conservatism somewhere around 2019 to 2022. These men are mostly young, mostly online native, mostly making money from audience capture. The interest served is economic and attentional. Anti-Jewish content performs on the algorithms that reward transgression. The Overton window on Jewish criticism has shifted enough in the past three years that you can now build a six-figure to seven-figure media career on it. Fuentes went from being banned everywhere to dining at Mar-a-Lago. The coalition does not need MacDonald’s theory in any developed form. It needs a rotating supply of incidents, ADL statements, Israel news cycles, and Jewish faces in elite positions to point at. The function is engagement. The interest is the audience. These men will mostly age out or get bored or move to whatever the next algorithmic transgression-niche is. While they are there they amplify the dissident-right intellectual product to audiences orders of magnitude larger than the intellectual product would otherwise reach.
The third is the Black nationalist tradition. Nation of Islam, the Hebrew Israelites in their various sects, the strain of Black Christian nationalism that sees Jews as the enemy of Black liberation, scattered figures like Kanye in his 2022 phase, the older tradition that runs through Farrakhan and back through the 1960s. This coalition’s anti-Jewish content has different content from the dissident right. The dissident right says Jews advanced Black interests against White interests. The Black nationalist tradition says Jews exploited Black labor, owned the slave ships, control the entertainment industry that degrades Black culture, and stand in the way of Black autonomy. The two coalitions agree on almost nothing except the conclusion. The interest served is internal to Black nationalism. Anti-Jewish framing lets Black nationalist leaders explain Black underperformance without conceding either to White-supremacy explanations or to internal-failure explanations. There is a third party. The third party is Jews. The function is the same as MacDonald’s function for the dissident right. It is moral cover for a coalition that needs an external enemy to maintain its solidarity.
The fourth is the post-October-7 left coalition. Pro-Palestinian student activists, DSA chapters, the Squad-aligned wing of the Democratic Party, the cultural left that has spent the past decade absorbing third-worldist frameworks, the academic humanities departments where Israel-as-settler-colonial-state is the consensus position. This coalition will resist the label anti-Jewish and the resistance is partly fair. The intellectual content is anti-Zionist, not anti-Jewish in the dissident-right sense. But the line between anti-Zionism and anti-Jewish content gets thin under pressure, and the pressure has been continuous since October 7. Mamdani’s elevation to the Democratic mayoral nomination in New York is the political end of this coalition. The interest served is coalition solidarity within the left. Anti-Israel framing unifies progressives across racial and class lines in a way that few other issues do. It also serves a generational function. The post-2020 left needed a moral cause that was not domestic identity politics, because domestic identity politics had become exhausted and electorally toxic. Palestine provided the cause. The function is unifying and energizing the post-Bernie left.
The fifth is the conspiracist-populist mass coalition. Alex Jones, RFK Jr.’s adjacent supporters, the QAnon residual, the Joe Rogan ecosystem when it tilts that way, the broad paranoid populism that does not have a coherent ideology but that periodically locks onto Jewish targets. Soros conspiracies, World Economic Forum conspiracies, central banker conspiracies. This coalition is the largest by raw numbers and the least organized. Its anti-Jewish content is mostly displaced. The conscious target is globalists, elites, the deep state, the cabal. Jewish names get attached because Jews are overrepresented in finance and Hollywood and so the conspiracy lands on Jewish faces by gravity rather than by design. The interest served is the explanation of unfair-feeling outcomes through agency rather than structure. People whose lives have gotten worse over thirty years need someone to be responsible. The coalition gives them the cabal. Sometimes the cabal is Jews, sometimes it is not, but the cabal is always small and always intentional. The function is the same as the function of all populist conspiracism since the Populist Party of the 1890s. Coherent agency feels better than incoherent structure.
These five coalitions interact but they do not unify. The dissident right intellectuals despise the streamers as cruder versions of themselves. The streamers see the intellectuals as boring. Black nationalists have nothing to do with either. The post-October-7 left would burn its own hand off before it would acknowledge alignment with the dissident right, and the dissident right is correspondingly hostile to the post-October-7 left’s third-worldism. The conspiracist-populist mass coalition floats above all of them and gets recruited by whichever of the others can attach itself successfully.
What unifies them at the highest level of abstraction is that all five face a coalition-formation problem to which Jewish targeting offers a partial solution. The dissident right needs an enemy that explains White political weakness without conceding moral inferiority. The streamers need transgressive content that performs. Black nationalism needs a third party that explains Black outcomes without conceding to either pole of the standard race debate. The post-October-7 left needs a unifying moral cause for a fragmenting progressive coalition. The conspiracist-populist mass needs agency behind incoherent structural change.
Notice the pattern. Each coalition has a problem that does not actually require Jewish targeting in principle. Each could solve its problem with a different enemy. The dissident right could blame Whites who collaborate, and sometimes does. Black nationalism could blame Asians who occupy similar middleman positions, and sometimes does. The post-October-7 left could focus on Saudi or Emirati or Egyptian state behavior, which is also bad, and largely does not. The conspiracist-populist mass could blame the Chinese Communist Party, and sometimes does. Why does Jewish targeting nonetheless emerge from each of these coalitions at high frequency?
Because Jews have four properties that no other available target combines. They are a market-dominant minority in the Chua sense, conspicuously successful in finance, media, law, medicine, academia, and politics. They are small enough to plausibly defeat. They have a state that takes recognizable nationalist actions, which provides a rotating supply of news cycles. And they have a defensive infrastructure, the ADL, AIPAC, the major Jewish federations, that responds to criticism in ways that confirm the coalitions’ frame. Every ADL letter and every Israeli military operation refreshes the supply of grievances that each coalition can use. The coalitions do not need to coordinate. The supply renews on its own.
The interests served are accordingly five different interests. The dissident right gets moral cover. The streamers get money and attention. Black nationalism gets internal coherence. The post-October-7 left gets unifying energy. The conspiracist-populist mass gets agency. Five problems, one available solution, no coordination required.
The American Jewish community, which is itself not one coalition but several, faces the unusual position of being targeted from five directions at once for five different reasons. Reform Jewish leaders trying to respond to dissident-right criticism from the right cannot use the same arguments they would use against post-October-7 criticism from the left. The Orthodox community has aligned with the right in ways that infuriate the Reform leadership. AIPAC has aligned with whichever party will support Israel and now finds itself defending Republican administrations its donor base mostly opposes domestically. Each Jewish sub-coalition is making its own bet about which threat is more dangerous and which alliance is more reliable, and the bets are diverging. There is no unified Jewish response because there is no unified Jewish coalition.
The question of which of these five coalitions is most dangerous to Jews is genuinely open and Jews disagree about it. The ADL and most of the legacy Jewish establishment treat the dissident right and its streamer amplifiers as the primary threat. The Orthodox community and the Israeli-aligned organizations increasingly treat the post-October-7 left as the primary threat because the left is closer to actual political power on questions about Israel. The smartest analysts in the Jewish community treat the conspiracist-populist mass as the most dangerous in the long run because its size dwarfs the others and because in conditions of economic crisis it can be activated by whichever of the smaller coalitions is best positioned at that moment.
The honest summary is that anti-Jewish discourse in America today is not one phenomenon and serves not one set of interests. It is the surface signature of five different coalition-formation problems that happen to point at the same target for structurally similar but substantively different reasons. Treating them as one phenomenon flatters the dissident right by inflating its importance and flatters the post-October-7 left by letting it deny any connection to the dissident right. Both kinds of flattery should be resisted. The wars are real. So is what the combatants are fighting about. The combatants are not the Jews. The combatants are the five coalitions, each fighting a different war, each finding the same target useful, none of them coordinating with the others, all of them feeding off the same renewable supply of grievances that the target itself, through no unified intention, continues to generate.
Jared Taylor is the most interesting figure in this whole landscape because he is the test case for whether the dissident right could have been built without anti-Jewish content. The answer is no, and Taylor’s career proves it.
Taylor founded American Renaissance in 1990 with an explicit policy that the Jewish question was off the table. The 1994 inaugural conference had four Jewish speakers out of ten, including Michael Levin and Mayer Schiller. Levin’s book is still sold by AmRen. Taylor said for years that Jews were White, that they should be welcome in any pro-White movement, and that obsession with Jews was a distraction from the real work of advocating for European Americans. He maintained this position with admirable consistency through the 1990s and into the 2000s. His personal style, the bow ties, the Yale education, the soft Southern manners, the willingness to debate anyone politely, was designed to demonstrate that race realism could be respectable.
What happened next is the data point. The movement Taylor was trying to build kept getting captured by the people he was trying to exclude. By the late 2000s the AmRen conferences had become sites where, as the Pyke article in MacDonald’s own Occidental Quarterly admitted, the Jewish question surfaced in almost every speech. Taylor’s Jewish supporters started leaving. Auster left. Levin distanced himself. The eleventh-chair article was MacDonald’s wing telling Taylor that his coalition rules did not work, that you could not build a White advocacy movement that included Jews, that the supposed allies were really competitors. The dissident right intellectual coalition I described in the previous answer was the wing that won that internal fight. Taylor’s wing lost.
Taylor himself never converted. He still does not publish anti-Jewish material. He still appears on platforms with Jewish guests. He still maintains the official AmRen line that Jews are White and welcome. But the people he is now in coalition with, Spencer at the height of the alt-right, Fuentes since, the various streamers, and MacDonald himself with whom Taylor has appeared repeatedly, are uniformly hostile to that line. Taylor’s neutrality has become a kind of personal eccentricity within a movement that has moved past him. He gets to be the respectable face. The respectable face is still attached to the body, and the body believes things he claims not to.
That Cofnas sent Taylor the 2018 paper and Taylor ignored it, is the diagnostic. Cofnas was offering Taylor exactly the ammunition Taylor needed to defend his original position. The 2018 paper demolishes the scholarly basis for the claim that Taylor’s allies use to justify excluding Jews from White advocacy. If Taylor had taken Cofnas’s paper seriously, he could have used it to push back against the Fuentes-Spencer-MacDonald wing and reassert the original AmRen line. He did not. The silence is informative.
Several explanations are available and they are not mutually exclusive. The first is that Taylor cannot afford to alienate the wing that now provides his audience, his speakers, and his donor base. AmRen in 1994 had a different reader profile than AmRen in 2024. The 1994 readers tolerated the no-Jewish-question rule because they wanted respectability. The 2024 readers want anti-Jewish content and tolerate Taylor’s personal restraint as an idiosyncrasy. If Taylor publicly endorsed Cofnas’s demolition, his current readers would leave. He would be left with a coalition of one. The second explanation is that Taylor has privately moved closer to the MacDonald position than he is willing to say. He has co-authored or co-appeared with MacDonald, has not publicly criticized him, has hosted speakers who endorse MacDonald, and has watched MacDonald’s framework become the dominant frame in his own movement without objecting. Conduct over thirty years suggests at minimum that Taylor finds the MacDonald position tolerable even if he does not adopt it personally. The third explanation is that Taylor is a tactician who decided that the official no-Jewish-question rule was always pragmatic rather than principled. He thought it would let the movement recruit more broadly. When that bet failed and the movement recruited better with anti-Jewish content than without, Taylor adjusted. He kept the personal rule for branding reasons but stopped enforcing it as a coalition rule. The fourth explanation is the one Taylor would offer if asked. He is committed to White advocacy, the Jewish question is a distraction, but he is not going to publicly attack his own allies over an internal dispute. This is the gentleman’s code. You do not air the family’s disputes in public. You handle them privately or not at all.
The four explanations together describe what coalition leaders actually do when their coalition shifts under them. They do not publicly resist the shift, because public resistance costs them the coalition. They do not publicly endorse the shift either, because endorsement costs them the personal brand they built before the shift. They go silent. The silence reads as complicity to outside observers and as restraint to insiders. Both readings are correct. Taylor is complicit and restrained simultaneously. The two are not in tension when you understand that coalition position requires both.
What Taylor’s career demonstrates is that the dissident right could not in fact be built without anti-Jewish content. Taylor tried for fifteen years. He failed. The movement he built kept generating the content he tried to exclude, kept attracting the personnel he tried to keep out, and kept rewarding the framers, MacDonald above all, whose work he claimed not to need. The coalition has a problem that requires a unified external enemy to solve. Whites who have no enemies are just Whites, and Whites who are just Whites do not need a movement. The movement requires the enemy. Taylor wanted a movement without an enemy and got a movement with the enemy he tried to exclude.
There is a smaller and more interesting parallel here. The Republican establishment between roughly 1955 and 2016 also tried to maintain a coalition that included anti-Jewish elements without endorsing them. Buckley famously read the John Birch Society out of the conservative movement in 1962 and tried to do the same thing with various anti-Jewish figures over the years. Buckley succeeded for a few decades because the conservative coalition during the Cold War had a different enemy, communism, that did most of the unifying work. When the Cold War ended the unifying enemy disappeared and the coalition fragmented along exactly the lines Buckley had tried to suppress. By 2016 the Buckley filtration system had collapsed and Trump rebuilt the Republican coalition on a different basis that did not exclude the elements Buckley had excluded. Taylor’s situation is the same dynamic at a smaller scale and within a more extreme movement. The moderate gatekeeper holds the line as long as the coalition has another enemy to do the unifying work. When the alternative enemy weakens or fails, the coalition reverts to the original enemy.
For Taylor specifically the original enemy was always going to be Jews because the dissident right was always defining itself against an elite it perceived as anti-White, and the most visible faces of that elite, in media and academia and finance and law, are disproportionately Jewish. Taylor wanted to attack the elite without naming the most visible faces. His allies wanted to attack the most visible faces. The most visible faces won the internal argument because they were more concrete than Taylor’s abstraction. Coalition formation rewards concreteness. Taylor lost on a structural feature of how coalitions form, not on the merits of his position.
Cofnas’s 2018 paper is correct on the merits. Taylor’s silence in response is also correct on coalition logic. Both can be true. The wars are real. So is what the combatants are fighting about. The combatants are not always who they say they are, and what they are fighting about is sometimes not what they say it is. Taylor knows he cannot use Cofnas’s paper because using it would dissolve his coalition. Cofnas knows Taylor cannot use it for the same reason. Cofnas sent it anyway, perhaps as a gesture of respect, perhaps as a test, perhaps as documentation. Taylor’s filing it without response is the diagnostic finding. The man who built his career on the proposition that the dissident right could exist without anti-Jewish content has spent the last fifteen years watching his own movement falsify the proposition, and his silence is what falsification looks like when the falsified party cannot afford to acknowledge the falsification.
Why hasn’t anyone in the AR seriously engaged Cofnas’s work on MacDonald? Because engagement would cost more than silence and yield less. The cost-benefit runs the wrong direction for everyone in the AR who could plausibly do it.

Start with the population of people on the dissident right who are intellectually equipped to engage Cofnas seriously. The list is small. MacDonald himself, Andrew Joyce, Edward Dutton, maybe Ricardo Duchesne, maybe Gregory Hood, maybe a few academics writing under pseudonyms at Counter-Currents or the Occidental Quarterly. Below that level the work would be sub-scholarly and would not register as engagement. The serious engagers number maybe ten people worldwide. Of those ten, the ones who depend on MacDonald’s framework as the load-bearing wall of their political worldview cannot engage Cofnas without risking the wall. The ones who do not depend on it have no incentive to spend the months it would take to write a serious response, because the response would alienate the readers who do depend on it.

MacDonald himself wrote the 2018 reply, the 2022 reply that got retracted, and the various blog posts. The replies are weak. Cofnas catches them being weak. MacDonald keeps replying because he has to, but he is not capable of producing a stronger reply because the underlying scholarship will not support a stronger reply. The errors Cofnas documents are not errors of interpretation that could be redescribed. They are sourcing errors of the form “this page does not contain the claim attributed to it,” which cannot be fixed by reframing. MacDonald’s only available moves are concession on small points, deflection to other topics, and assertion that Cofnas misunderstands the larger framework. He has played all three. None of them work. He has nothing left.

Joyce is the second most plausible engager. He has written prolifically for the Occidental Observer and produced book-length material on Jewish topics. Joyce’s work has the same structural problem as MacDonald’s, which is that it relies on selecting Jewish examples and treating the selection as evidence of a unified pattern. A serious engagement with Cofnas would require Joyce to defend the selection methodology, which he cannot defend, because the methodology is the problem. So Joyce has stayed at the level of producing more examples rather than defending the framework. Producing more examples is what the readership wants anyway. Defending the framework against Cofnas would not produce more examples. It would just absorb time that could be spent producing examples.

Dutton wrote a defense of MacDonald in 2018 that Cofnas demolished in a 2019 reply published in Evolutionary Psychological Science. Dutton’s defense made an elementary statistical error about Jewish in-marriage rates that Cofnas caught immediately. After that exchange Dutton mostly stopped engaging on the topic. He has written on adjacent topics but has not returned to a serious defense of MacDonald against Cofnas. The Dutton case is informative because it shows what happens when someone in the AR orbit does try to engage at an academic level. They get embarrassed and withdraw. The withdrawal is rational. Other potential engagers watched what happened to Dutton and drew the obvious conclusion.

Below this top tier the engagers are not capable of academic-level response. They are bloggers, podcasters, or polemicists. Their readers do not want academic-level response. Their readers want polemic. So the apparent engagement at the lower levels takes the form of articles with titles like “Cofnas refuted” or “the controlled opposition exposed” that do not actually address Cofnas’s specific sourcing claims but instead attack his motives, his employer at the time of writing, his Jewish background, his subsequent departure, his alleged opportunism. This is engagement of a kind. It is not engagement of the kind that would settle anything. It is engagement that performs loyalty to the coalition while leaving the scholarly question untouched.

Then there is the structural problem. Cofnas is Jewish. Any serious engagement with him by an AR figure becomes vulnerable to the move “you are debating a Jew about Jews, which is the trap they want you in.” This move is silly but it works on the audience. The audience does not want to see its leaders treating a Jewish critic as a peer. Treating Cofnas as a peer would suggest that scholarly debate with Jews is possible and productive, which is the proposition the coalition is built on denying. So engagement with Cofnas at the level of peer-to-peer scholarly exchange is structurally forbidden by the coalition’s own rules. You can attack Cofnas but you cannot debate him. MacDonald did debate him, in the 2018 and 2022 replies, and those replies are exactly the moments where MacDonald’s coalition position weakened. The lesson the rest of the AR drew is that you do not give Cofnas the dignity of response.

There is a third reason which is more interesting. Cofnas’s later trajectory gave the AR an out. After the 2018 to 2023 papers, Cofnas moved into hereditarian race-and-IQ work, lost his fellowship in 2024, became a cause célèbre on the heterodox right, and started writing material that sounds increasingly like material the AR itself would publish if it could find an academic with credentials. The AR’s response to this trajectory has not been to revisit his MacDonald work. The response has been to absorb him as a partial ally on race and IQ while continuing to ignore his MacDonald work. He gets cited approvingly on hereditarianism. He gets ignored on MacDonald. This is coalition behavior at its most efficient. Take what you can use, ignore what you cannot, do not let the parts touch each other.

Cofnas himself has not pressed the issue. He has moved on to other projects. He could have spent his career hammering on MacDonald and become the recognized academic authority who finally killed MacDonaldism. He chose not to.

There is a fourth reason, the deepest one. Even an AR figure who privately agreed with Cofnas on every point would not say so publicly because saying so would not change the readership’s behavior. The readership does not read MacDonald because it has been convinced by the scholarship. It reads MacDonald because the framework provides what we discussed earlier, moral cover for white solidarity, emotional integration of real grievances, comprehensive narrative of coalition competition. None of these functions depend on the scholarship being true. So even a public AR endorsement of Cofnas would not dissolve the readership’s commitment to MacDonald. It would just lose the endorser the readership. The endorser would be giving up the audience without buying anything in return. No rational coalition leader makes that trade.

Compare to what would have to be true for serious engagement to happen. An AR figure would have to be intellectually capable of engaging at academic level, willing to lose readers who are committed to MacDonald, willing to be accused of debating a Jew about Jews, confident that the engagement would produce better results for the coalition than continued silence, and personally indifferent to the social cost within the AR of being seen as the one who broke ranks. The number of people who satisfy all five conditions is approximately zero.

Taylor is the closest. Taylor has the intellectual capacity, has the prestige to absorb the social cost, has demonstrated for thirty years that he is willing to take the no-Jewish-question position, and would benefit politically from Cofnas’s argument because it vindicates Taylor’s original AmRen line. Taylor still has not engaged. The previous answer covered why. Taylor’s silence is the strongest possible evidence that engagement is structurally impossible even for the AR figure who would benefit most from it. If Taylor will not do it, no one will.

Spencer is an interesting subcase. Spencer has the intelligence, has the credentials adjacent enough, and has periodically broken from MacDonaldism in interviews. After his 2017 collapse he became a strange kind of moderate within the dissident right, criticizing some forms of explicit anti-Jewish content while continuing to hold racialist positions. Spencer has the option of engaging Cofnas seriously. He has not. The reason in his case is that Spencer no longer has a coalition to lead. He is a discredited figure within his own movement. Engagement with Cofnas would not rebuild his coalition. It would just produce another podcast appearance. The marginal value is too low.

Below Taylor and Spencer the field thins out fast. The men who would have been positioned to engage twenty years ago are mostly dead, retired, or have moved into other projects. The new generation is podcasters and streamers who do not engage at academic level on principle. Fuentes will not engage Cofnas because Fuentes does not engage anyone at that level. The Daily Wire will not engage because the Daily Wire is officially philo-Semitic. The mainstream conservative outlets will not engage because they treat MacDonald as outside the bounds of serious discussion and engaging would dignify him. Cofnas’s work falls into a gap between conservative outlets that consider MacDonald beneath notice and AR outlets that consider Cofnas a hostile actor whose work cannot be addressed without lending him legitimacy.

The result is that the most thorough academic demolition of the dissident right’s intellectual founder sits in three peer-reviewed articles that almost no one in the dissident right has read, that the few who have read them cannot publicly engage, and that the broader public never hears about because the mainstream considers the underlying question too distasteful to elevate. Cofnas is right and ignored. MacDonald is wrong and influential. The asymmetry is sustained by exactly the coalition mechanics that the original work failed to name. Cofnas refuted MacDonald on the scholarship and lost on the coalition. He could not have won on the coalition because he was not building one. MacDonald was building one. He still is. That is why the work regenerates and why no one engages the demolition.

There is one final wrinkle. Some AR figures privately read Cofnas and find his arguments persuasive, and adjust their own work accordingly without crediting him. This is the most common form of actual engagement. You can see traces in the more careful AR writers who have quietly stopped making the specific claims Cofnas demolished and shifted to claims he did not address. This is not citation. It is absorption. The AR has absorbed the parts of Cofnas’s critique it cannot rebut, while continuing to deny that absorption has happened. This is also coalition behavior at its most efficient. Update silently, never give the critic credit, keep the framework intact at the surface level while quietly improving the underlying claims.

The complete picture is that no one engages because everyone who could is either too committed to the framework, too dependent on the readership, too vulnerable to coalition discipline, too discredited to recover from engagement, or too sensible to spend the time. The work sits there. It will keep sitting there. Twenty years from now someone in the AR will rediscover Cofnas’s papers and treat them as new objections, and the same dynamic will repeat. The wars are real. The combatants do not always engage. Sometimes the most effective response to a devastating critique is to refuse to acknowledge that the critique exists. The AR has refused successfully for seven years. It will refuse for as long as the coalition holds, which is to say for as long as the underlying conditions that produced the coalition continue to hold, which is to say for the foreseeable future.

‘Analyzing Kevin MacDonald’s ‘Culture of Critique’ and the alt-right’s embrace of anti-Jewish ideology’ (April 16, 2018)

Cofnas’s strongest move is the both-sides evidence. If Jewish overrepresentation in twentieth-century intellectual movements ran on the engine MacDonald describes, high IQ plus ethnocentrism deployed against gentile interests, you should not see Jews leading the BDS movement, defending hereditarianism, founding FIRE, or critiquing psychoanalysis. The pattern Cofnas documents fits a supply-side story, where a high-IQ urban population produces leaders across the spectrum, better than a coalition-purpose story.
MacDonald’s reply concedes the game. Once he shifts from headcount to influence, the theory loses its grip on falsifiability. Whatever side wins gets read backward as the Jewish side. Lakatos called this methodological degeneration: weakening the theory to absorb anomalies without making new risky predictions. The intermarriage move is the same problem in miniature. A seventy percent outmarriage rate is the opposite of what a group evolutionary strategy predicts. Calling outmarriage part of the strategy turns the theory into a sponge.
Where Cofnas is thinner: his default hypothesis treats Jewish political tilt as pure cognitive supply. That underexplains things. Most ethnic groups show average political slopes that come from history, not conspiracy and not pure IQ. Second-generation immigrant Jews in early twentieth-century New York had reasons to find Boas more congenial than Madison Grant that did not require ethnic strategy and did not reduce to test scores. Cofnas wins against MacDonald by collapsing the question into supply, but the truth probably sits between supply and average group tilt.
The most useful point in the essay is the last one. When mainstream institutions deny easy facts, such as Jewish overrepresentation in elite positions or average group differences in measured traits, they create the market for the people who will say what is denied. The audience reasons: if they lied to me about the easy things, why trust them on the hard ones. MacDonald’s growth depends on that distrust more than on his evidence. David Reich’s admission that mainstream geneticists have masked the possibility of group differences for political reasons hands MacDonald his recruiting line.
From the Alliance Theory angle, both men occupy coalition positions. MacDonald writes for venues read mostly by White nationalists. His income, status, and protection come from an audience that rewards finding Jewish influence everywhere. Cofnas writes for the heterodox-respectable right around Pinker, Reich, and Murray. His coalition rewards him for being tough enough on IQ science to stay credible while drawing a clean line against the conspiratorial extension of that science. Neither incentive proves either man wrong. Both incentives shape what each man will and will not see.
The Lakatos point is the cleanest takedown. The both-sides evidence is the cleanest empirical point. The noble-lie-backfire point is the most useful for understanding why MacDonald has readers at all. Cofnas wrote a careful essay that does the disassembly without the usual moralizing, which is rare for the topic and part of why it landed.

‘Power in Cultural Evolution and the Spread of Prosocial Norms‘ (Dec. 2018)

Cofnas argues that Boyd-Richerson cultural evolution models miss the obvious. Hunter-gatherer morality did not aggregate from individual learning biases. Enforcement came from a coalition of the rank-and-file against would-be alphas and against anyone else who broke ranks. He leans on Boehm’s reverse dominance hierarchy to argue that the first coalition was the coalition of subordinates, and that morality and coalition enforcement arose together, not in sequence.
The Jurisdictional Wars logic does not start with sedentarism or with the founding of states. It starts when a band of subordinate males formed the first majority coalition and used collective force to control the alphas and punish anyone who failed to share meat or violated taboos. Moral vocabulary and coalition enforcement are the same phenomenon from the beginning. You could read the Cephu story as a small-scale jurisdictional war. Cephu makes a sovereignty claim, calling himself a chief of his own band. Kenge’s coalition denies the category. The Mbuti do not have chiefs. If Cephu wants to be a chief, he can leave the forest, which means die. The response is jurisdictional. The coalition does not recognize his standing, and the moral vocabulary (“Animal!”) does the boundary work.
Where Cofnas flinches is in his treatment of the coalition of the majority as if it were a unitary actor with shared foresight. He writes of “blueprints” the group “agrees on” and norms enforced “with explicit awareness of the social benefits.” Pinsof might push back. Within the coalition of the majority there are sub-coalitions, and whoever wins the internal politics gets to define the moral vocabulary the larger coalition then enforces. Which men shape the consensus? Whose definition of fairness becomes the rule? Cofnas does not ask. He treats “the group” as the agent and “group benefit” as the criterion. That is the same essentialism Turner attacks in Vermeule and Deneen, relocated to the Paleolithic.
Functionalism creates the related problem. Cofnas needs the norms to be group-beneficial because the alternative undercuts his deliberate engineering story. The alternative: coalitions enforce vocabularies whose consequences nobody predicts and whose benefits flow asymmetrically inside the coalition. Chudek’s endocannibalism example sits awkwardly here. Whose group benefit did that serve? An Alliance Theory angle reading does not need the norms to be functional for the group. It needs them to mark coalition membership and to give the enforcers something to enforce.
The concession to cultural evolutionary modeling for “nonmoral” domains is also worth pressing. Cofnas grants Boyd-Richerson the territory of spear-making technique and other practices the coalition treats as private. The moral and nonmoral line is not stable. Almost anything inside an elite institution becomes moralized once it enters coalition politics. Consider DEI hiring, peer review, citation practice, manuscript style. None of it was supposed to be moral. All of it is now. Cofnas treats the moral domain as a bounded category. The Jurisdictional Wars framework treats moralization as a process that can absorb any cultural domain a coalition decides to police.

‘Should dietary guidelines recommend low red meat intake?’ (Sep. 5, 2019)

Cofnas co-writes a defense of meat against the EAT-Lancet recommendations.
The authors make a defensible case on the narrow scientific question. Nutritional epidemiology relies on food frequency questionnaires that people fill out poorly, produces relative risks under 2 that would not pass muster in most other epidemiological fields, and rarely survives translation to randomized controlled trials. Ioannidis has been making this point for years across the whole field. The hazelnut example lands. If you take the meta-analyzed cohort data at face value, a daily hazelnut adds a year of life. Nobody believes that, and yet the same methodology produces the meat numbers that the EAT-Lancet Commission treated as decisive. The authors are right that the case for severe meat restriction does not clear the evidentiary bar that would be required in almost any other domain.
The Jurisdictional Wars reading sits one layer above the science. EAT-Lancet is a coalition product. The Commission combines climate scientists, nutrition researchers, foundation funders, and animal-welfare advocates whose moral commitments to plant-based eating predate the evidence. Leroy and Cofnas note the Seventh-day Adventist line running through nutrition guidance, citing the Banta paper on Adventist influence on global dietary advice. Adventist health messaging has shaped American nutrition orthodoxy for over a century, from Kellogg through Loma Linda, and the moral vocabulary of the modern plant-based movement carries Adventist DNA whether the speakers know it or not.
Who does an EAT-Lancet commissioner rely on for status, income, and protection? Foundations, journals, and a public health establishment that has staked its credibility on the meat-is-bad consensus. Who must they retain as allies? Climate funders, animal-welfare donors, the journals that publish the cohort studies, the activist NGOs that amplify the conclusions. What signals coalition membership? Citing the right cohort studies, treating relative risks under 2 as decisive when meat is the exposure, framing dissent as industry-funded. What would they have to give up to change position? Tenure cases, grant renewals, the moral standing of having been on the right side. Once you ask those questions, the asymmetry in how the evidence gets handled becomes legible. Visceral fat shows a relative risk of 5.9 for colorectal neoplasia and gets modest attention. Red meat shows under 1.2 and becomes the centerpiece of a planetary diet recommendation. The science does not explain the asymmetry. The coalition does.
Becker’s hero system frames the planetary diet language. EAT-Lancet calls for a Great Food Transformation, sanctifies a reference diet, and proposes warning labels and taxes. That vocabulary marks dietary virtue as the path to immortality through a saved planet. Meat eaters become the obstacles to collective salvation. The structure is religious in the sense Becker meant. The policy prescriptions follow from the moral architecture, not from the relative risks.
The commissioners do not experience their position as coalition signaling. They experience it as following the evidence. The cohort studies feel decisive because the moral conclusion is already settled, and the methodological skepticism that researchers apply elsewhere goes quiet when the conclusion aligns with the coalition’s prior commitments. Schoenfeld and Ioannidis showed that 40 of 50 cookbook ingredients had been associated with cancer risk in observational studies. Researchers know this. They apply the discount to other foods. They stop applying it when the food is meat.
Turner’s anti-essentialism applies to the category of red meat itself. EAT-Lancet treats red meat as a stable kind with intrinsic health properties. The category collapses on inspection. Beef from a feedlot, beef from grass, processed lunch meat, fresh cuts, organ meats, the meat in a hunter-gatherer diet that Cordain documents as two-thirds or more of caloric intake without cardiometabolic disease — these get bundled into one moral object. The bundling is what lets the coalition treat a heterogeneous food group as a single villain.
Leroy has received funding from meat industry sources, which the paper does not declare in the way an EAT-Lancet conflict-of-interest section would. That does not make his methodological points wrong, since the same critique of nutritional epidemiology comes from Ioannidis and from researchers with no meat-industry ties. But the symmetry cuts both ways. Industry money shapes the meat-defense literature the way activist money shapes the plant-based literature, and a careful reader keeps both in view.
The hunter-gatherer evidence the paper cites is real but does less work than the authors imply. Cordain’s numbers on animal-source caloric percentages are contested, and the cardiometabolic profile of foragers reflects total lifestyle, not meat in isolation. The paper leans on this evidence because it suits the argument, which is the same move they accuse EAT-Lancet of making with the cohort data. Both coalitions reach for the evidence that confirms and discount the evidence that complicates.
The paper’s topic is an example of how a contested empirical question gets resolved by the coalition that wins the institutional fight rather than by the evidence. The science does not settle the meat question. The coalition does. EAT-Lancet won the journals, the WHO, the public health establishment, and the dietary guidelines. Leroy and Cofnas are writing from outside that coalition, which is why their methodological points, defensible on the merits, will not move policy.

‘A debunking explanation for moral progress’ (Oct. 30, 2019)

Cofnas builds a debunking case for moral progress that runs through self-interest plus empathy. The argument has force in places and breaks down in others.
The strong parts come first. Cofnas rightly notes that much liberalization runs on self-interest. Slaves wanted freedom. Peasants wanted protection from lords. Women wanted property rights. No one needs moral realism to explain any of that. The Hobbesian trap account of hunter-gatherer warfare also works. People dislike living in fear, but coordination problems trap them in violence until a central authority breaks the cycle. Pinker’s data on declining homicide rates supports this picture.
Boehm’s reverse dominance hierarchy thesis carries the paper’s anthropology. The idea that hunter-gatherer bands suppressed alphas through coalitions of subordinate males maps onto Alliance Theory angle. Coalitions of the weak constrain the strong. That logic recurs across history.
The weak parts come faster than Cofnas seems to notice.
First, he leans on Sapolsky for the empathy-as-evolved-trait argument. Sapolsky overstates and overreaches throughout Behave. The kin selection and reciprocal altruism story for empathy is real evolutionary biology, but Sapolsky’s framing carries ideological baggage that Cofnas absorbs uncritically.
Second, the empathy debunking cuts deeper than Cofnas wants. If our moral cognition evolved for fitness rather than truth, the same skepticism applies to all our cognitive faculties, including the reasoning Cofnas uses to build the argument. He notes the problem in passing and moves on.
Third, the self-interest account works for in-group liberalism but strains for expanding moral concern to those with no political clout. Animal welfare, foreign aid debates, concern for future generations, treatment of the disabled. None of that runs on the self-interest of the beneficiary, since the beneficiary cannot enforce anything. Cofnas concedes the point about animals and lets it pass. Empathy alone seems thin. A coalition account might do better. Caring for the powerless marks membership in the dominant moral coalition. Status accrues to those who display the right concern. The beneficiary’s lack of power becomes the point, not a problem.
Fourth, the account has a Whiggish shape. Cofnas treats liberalization as the dominant trend, with reversals as noise. But the populist resurgence, illiberal movements across Europe and India, the retreat from free trade, the realignment around national interest, all suggest the trend was contingent on conditions that may not hold. If liberalism was a byproduct of postwar prosperity, secure borders, and pacification, what happens when those conditions break?
Fifth, the women’s liberation section is the weakest. Cofnas writes that men used their strength to arrange society for themselves but also cared about women, and so heeded women’s demands. This explains nothing. Men cared about women in 1300 too and arranged society for themselves anyway. The actual story almost certainly turns on industrialization changing the economic value of female labor, fertility transitions changing household economics, and contraception changing the male-female bargaining position. None of that is empathy. A coalition account might say that as women became economically independent, their alliance value to men shifted, and the male-female compact got renegotiated.
Sixth, Cofnas barely addresses why liberalism spread from particular cultural sources rather than emerging spontaneously everywhere. Larry Siedentop’s Inventing the Individual makes a strong case that Christian theology supplied the conceptual material for moral equality and individual dignity. Liberalism is not a generic outcome of pacification. It is a specific cultural product of Latin Christendom that conquered rival moral systems through technological and economic dominance. Cofnas’s universalist framing hides this.
Seventh, the metaethical conclusion is weak. Cofnas argues that since both realism and antirealism predict convergence, convergence cannot decide between them. Then he claims his naturalistic account is “superior.” Superior how? He gestures at parsimony but does not earn it. The realist can run the same parsimony argument in reverse.
Cofnas takes the empirical case seriously where many philosophers wave their hands but the framework needs the coalition layer to do the work he wants self-interest and empathy to do alone. Self-interest explains why oppressed groups push back. It does not explain why winning coalitions adopt particular moral vocabularies that signal membership and exclude rivals. That is where Alliance Theory angle does heavier lifting than what Cofnas reaches for.
He wrote this in 2019 from Balliol College, Oxford. Who does he rely on for status, income, and protection? The academic philosophy profession. Who does he need to attract? Editors at Philosophical Studies, supervisors, future hiring committees. What signals mark coalition membership? Citing the right philosophers, criticizing realism in a way that respects field stylistic norms, treating the question as live rather than settled. What would he give up if he changed position? Publication access, professional respectability, a tenure-track future. The paper sits comfortably within those constraints. That does not make it wrong, but it explains some of its evasions.

‘Coercive paternalism and the intelligence continuum’ (2020)

Cofnas accepts Conly’s case for coercive paternalism and tries to strengthen it by adding a premise she explicitly rejects: that intellectual capacity differs across people in ways that bear on who needs protecting from his own choices. The move is clean. Conly claims paternalism rests on shared fallibility, not superiority. Cofnas says no, the fallibility is unequally distributed, and your argument grows stronger if you admit it.
The empirical section pulls real weight. The National Adult Literacy Survey data hold up. Most adults cannot reliably read a bus schedule and figure out a wait time. A quarter cannot interpret an appointment slip. About forty percent cannot follow directions for taking medication on an empty stomach. Williams and Davis and Gottfredson have done the work, and the numbers are what they are. The point that functional literacy and general reasoning track each other, and that both predict health outcomes after controlling for income and education, has been established many times.
What strains is the leap from those numbers to his policy proposals. Cofnas suggests appointing supervisors to monitor what low-IQ patients buy at the grocery store and which restaurants they enter, with power to forcibly intervene. He suggests compelling some to live in environments where they can be monitored more closely. He suggests financial penalties for not notifying a health professional at medication time. None of this follows from the literacy data. What follows from the literacy data is that we should write simpler instructions, use pictographic medication labels, build automated reminders, and design systems that do not assume a college reading level. Most help for low-functional-literacy populations does not require coercion. The coercion proposals do other work.
His handling of disparate impact stumbles most. He notes that ethnic groups differ in mean IQ and bottom-NALS representation, then waves off the concern by saying paternalistic help is by definition a benefit, so disparate impact cannot be adverse. This begs the question. Whether the intervention helps is exactly what is contested. The historical record of state intervention into the lives of populations classified as low-functioning, from eugenic sterilization to residential schools to civil commitment, shows these interventions failing by their own stated metrics while serving the interests of those who administered them. Cofnas writes as if the paternalist coalition has no interests of its own.
Source quality is mixed. Gottfredson is a serious researcher whose health-IQ work holds up. Kanazawa is a different matter. The savanna-IQ interaction hypothesis sits poorly with the rest of evolutionary psychology, and Kanazawa’s claims about liberalism, atheism, and obesity have not aged well. Building a chunk of the argument on him weakens the paper.
The deepest problem is essentialist. Cofnas treats low IQ as the relevant essence that licenses intervention. Turner’s whole point against this kind of move is that essentialist categories smuggle normative work in under descriptive cover. “Low IQ person” sounds like a description. In Cofnas’s argument it functions as a license to override autonomy. The g factor is real and predictive. The step from a real predictive variable to a category that confers reduced moral standing is not a small one, and Cofnas does not work for it.
He also passes over the strongest version of Mill. Cofnas treats the harm-to-others versus harm-to-self distinction as resting only on Enlightenment optimism about Homo economicus, and once that picture falls, the distinction falls with it. But there is a separate argument: even if I make bad self-regarding choices, the cost falls on me, and I have a stronger claim to make my own bad choices than I have to inflict bad choices on others. That argument does not require believing I am always right. It requires believing my life is mine in a way other people’s lives are not. Cofnas does not engage this version, which is the version that does most of the work in liberal political theory.
Worth running my four questions on the paternalist himself. Who administers a regime of supervisors monitoring grocery purchases and assessing dietary compliance? Social workers, healthcare bureaucrats, family court judges, a new class of certified supervisors. The argument grows the coalition that decides who needs supervising. The people supervised gain no status, income, or protection. The supervisors gain all three. Conly has the same problem, but Cofnas makes it sharper because his proposed interventions are heavier and his target population gets defined by a measurable trait that already correlates with poverty and vulnerability.
The paper is honest in a way most academic work is not. Cofnas says out loud what many soft-paternalists half-believe but will not write down. That honesty makes the weak parts more visible than they would be in a more cautious version of the same argument.

‘Are moral norms rooted in instincts? The sibling incest taboo as a case study’ (Aug. 25, 2020)

Cofnas does cleaner work here than in the paternalism paper. The argument is narrow, the target is well-defined, and he sticks to it. He takes Westermarck’s theory of the sibling incest taboo, applies Williams’s representation problem and Wolf’s moralization problem, and shows that the theory cannot do what its proponents claim.
The structural argument lands. The Westermarck effect, if real, says I avoid sex with my childhood coresidents. The taboo says everyone avoid sex with siblings. These differ in two ways. The objects differ: childhood coresidents are not the same set as siblings. The scope differs: a personal aversion is not a third-party prohibition. Cofnas is right that no amount of asserting that the instinct gets “translated” or “formalized” into the taboo bridges this gap. Wilson’s hand-waving on this point has always been weak, and Cofnas is right to call it out.
The handling of Sesardic’s reply is the strongest section. Sesardic’s move is clever. He says it does not matter what objectively triggers the aversion, only what the person experiences as triggering it, and people will reach for sibling rather than childhood coresident because the former is the available cultural category. Cofnas’s reply is exactly right. Sibling is an available category only because we already have the social norms that make it one. The story explains the taboo by presupposing it. This is the same kind of circle Turner identifies in essentialist arguments generally. The category that supposedly explains the practice is itself constituted by the practice.
The vicarious disgust section also holds up. Cofnas reads Lieberman and Lobel carefully, and the read is damaging. Their Study 1 found no effect of coresidence duration on moral attitudes toward third-party peer sex. Their Study 2 found a small effect in a rank-ordering task where peer sex landed near speeding on the highway in moral severity, far from sibling sex. Treating this as evidence that personal aversion produces moral condemnation overstates what the data show. The May point about the 14-on-100 rating in Wheatley and Haidt deserves wider notice. A mean of 14 on a wrongness scale is not condemnation. It is mild distaste that the experiment has labeled wrongness.
The Haidt material throughout is handled with appropriate skepticism, which matches my view of him as untrustworthy. Cofnas does not lean on Haidt’s broader claims about disgust and morality. He treats the specific empirical findings on their merits and notes the publication-bias concerns Landy and Goodwin raise.
The recognition hypothesis section is where the paper grows weaker, and Cofnas seems to know it. Durham’s data are suggestive but thin. Twenty out of sixty cultures showing some version of a bad-stock argument is real evidence that some societies noticed inbreeding depression, but it does not establish that recognition produced the taboo rather than the other way around. People with an existing taboo will rationalize it, and bad-stock stories are an obvious rationalization. The Pope Gregory quote shows the Church offering this reason, but the Church had other reasons too, as Goody and Prinz argue and as Cofnas concedes when he mentions the Church’s interest in collecting estates from those who died without heirs. Schulz et al. on cousin marriage and Western individualism is a serious finding, but it tells us about consequences of the ban, not origins.
The four diagnostic questions cut against him here. Who establishes incest taboos in early societies? Elders, religious authorities, the men who control marriage exchange. What do they gain? Control over the marriage market, the right to direct young women to politically useful unions outside the family, the moral authority to enforce all this. The taboo serves the coalition of those who run exchange. The bad-stock rationale and the Westermarck instinct are both downstream of that coalitional fact. Cofnas does not run this analysis. He treats the question as if it were purely cognitive, as if the only candidates were “people felt disgusted” or “people noticed birth defects.” Lévi-Strauss had at least the structuralist version of the coalitional answer, that the taboo forces alliance-building exogamy, and Cofnas mentions it briefly before moving on.
The deepest point in the paper is the closing claim that representation and moralization problems are general. The gap between an instinct’s content and a moral norm’s content shows up everywhere. Kin selection gives me a disposition to favor my children. It does not give me a moral judgment that everyone should favor their children. That moral judgment is a coalitional product. It is a rule the group endorses because the group benefits from people raising their own children rather than free-riding. Street’s evolutionary debunking argument depends on collapsing this distinction, and Cofnas is right that the distinction will not collapse.
Alliance Theory angle says moral vocabularies are coalition technologies. The representation and moralization problems are exactly what you would expect if Pinsof is right. Instincts are individual-level. Moral norms are coalition-level. They do different work, run on different logics, and cannot be derived from each other by any simple translation. Cofnas’s conclusion is congruent with that frame. Where he says the cultural evolution of morality is “not tightly constrained by our biological endowment,” you might say the constraint runs through coalitional selection on shared vocabularies, not through individual instincts.
Cofnas treats the Israeli kibbutz data as decisive against Westermarck via Shor and Simchai’s reanalysis showing most coreared kibbutzniks felt indifference rather than aversion. That reanalysis has held up better than Shepher’s original claims. But Shor and Simchai’s social cohesion theory has its own problems, which Cofnas notes. It also has the same coalitional question lurking under it. Why would small nonvoluntary groups establish norms against intragroup sex? Because the coalition holding the group together has an interest in preventing dyadic bonds from fragmenting it. This is closer to the right kind of explanation, but Cofnas leaves it as a sociological observation rather than pursuing the coalitional logic.
The paper is honest, well-argued, and a useful corrective to confident evolutionary just-so stories about moral norms. The implicit conclusion supports skepticism of the kind of bio-essentialism that runs through a lot of evolutionary psychology. Where Westermarck wants morality to grow out of instinct, Cofnas shows the gap. Filling the gap requires coalitional theory, which Cofnas does not provide but which his argument leaves room for.

‘The Anti-Jewish Narrative’ (Jan. 5, 2021)

The 2021 paper sharpens the 2018 one and adds the move that does the most work: framing MacDonald not as a fringe figure to be ignored but as the mirror image of the mainstream race narrative. Cofnas wants to position himself as a third option, the genuine race realist who rejects both the white-racism narrative and the anti-Jewish narrative. That positioning is rhetorically effective and methodologically interesting, whatever you make of the underlying race-realist commitments.
The intermarriage point is the cleanest blow. MacDonald made one falsifiable demographic prediction. It failed. The Reform and unaffiliated intermarriage rates of 50% and 69% mean the population MacDonald identified as carrying out the group strategy is voluntarily ending its own genetic line within two generations. MacDonald’s response, that intermarriage is part of the strategy through alliance with Trump and Clinton families, is the kind of move Alliance Theory predicts a coalition’s founding text will make under pressure. Once any outcome confirms the theory, the theory has stopped being about the world and started being about the coalition.
The Hugh Davis Graham misuse is the strongest sourcing point. MacDonald cites Graham as agreeing that Jewish organizations were a necessary condition for the 1965 Immigration Act. Cofnas reads the surrounding pages and shows Graham saying the opposite. The quota system was already being circumvented by executive parole, the Bracero program, and Western Hemisphere exemptions. Truman and Eisenhower had paroled 700,000 refugees outside the quotas. By 1960, two-thirds of immigrants entered without quota numbers. Graham says abolition was an idea whose time had come, that the consequences were unintended, and that Celler was so disturbed by the collapse in European immigration that he tried to introduce corrective legislation. None of that survives in MacDonald’s selective quotation. This is the same pattern as the Sanford misreading in 2018: pull a passage that sounds supportive, drop the surrounding argument that flips the meaning.
The Sweden point lands hard. If Jewish influence is a necessary condition for liberal multiculturalism, the most extreme multicultural country in the West, with a Jewish population of 0.2%, becomes inexplicable. Germany under Merkel is the same problem. The theory predicts a correlation between Jewish presence and multicultural policy. The correlation is not there.
The Israel material does what the 2018 paper did and tightens it. Israel grants automatic citizenship under the Law of Return to anyone with one Jewish grandparent and their non-Jewish spouse and children. 400,000 Israelis are not considered Jewish by the Chief Rabbinate. Up to 86% of recent immigrants may not be halachically Jewish. Liberal American Jews, including Dershowitz, lobbied to bring 135,000 Ethiopians who have no genetic relation to other Jewish populations. The “racial purity for Israel” frame collapses against this record. MacDonald’s response, that Ethiopians are only 2% so don’t threaten the demographic status quo, is the gotcha Cofnas spends a paragraph savoring, because it’s the exact argument MacDonald would reject if a Jew made it about America.
The American Renaissance material is the most underused asset in the paper. Four of ten invited speakers at the 1994 founding conference were Jewish. Levin’s book is still sold by AmRen. Auster and Levin left as anti-Semitism crept in. Pyke in MacDonald’s own Occidental Quarterly admits Jews showed up wanting to be allies and got the eleventh-chair treatment. The structure here is pure coalition logic. White nationalists demand that Jews join their movement to prove they aren’t ethnocentric, exclude them when they try, then cite the exclusion as evidence of Jewish ethnocentrism. The move is unfalsifiable by design. Cofnas catches it but understates how cleanly it shows the coalition function of the theory.
What Cofnas still doesn’t quite say is that MacDonald’s entire framework rests on essentialism in Turner’s sense. He treats Jews as a unified actor with a unified strategy, then treats every disconfirming Jew as either a hidden activist, a defector, or a useful idiot. Drop the essentialism and you get the question Cofnas keeps circling but never names: which Jewish sub-coalitions back which movements, and why? Reform Jews in postwar New York worked one alliance structure. Orthodox Jews in Brooklyn work a different one. Israeli right-nationalists work a third. Soviet refusenik conservatives work a fourth. The “Jews” of MacDonald’s theory are an artifact of treating a coalition-internal disagreement as a unified actor’s hypocrisy.
One real weakness in Cofnas’s paper. The “default hypothesis” of high IQ plus urban concentration is parsimonious for explaining overrepresentation in cognitively demanding fields, but it underdescribes the political tilt. Saying “right-wing movements were disproportionately anti-Semitic so Jews skewed left” is true but circular for a period when much of the left was also anti-Semitic, particularly the Soviet and Polish communist parties Cofnas referenced in 2018. The fuller answer requires looking at which specific coalitions in which specific cities at which specific moments admitted Jews on equal terms. That’s the analysis MacDonald should have done and didn’t, and it’s the analysis Cofnas gestures at but doesn’t quite execute.
The framing here, “genuine realists have to fight a war on two fronts,” is Cofnas explicitly recruiting a coalition. He wants race-realist readers to abandon MacDonald without abandoning race realism. Whether you find that move legitimate depends on what you think of the underlying commitment. The internal critique of MacDonald is sound either way.
The case is closed on MacDonald as scholarship. The case stays open on why the book keeps regenerating among readers who have access to the demolition. That’s a coalition question, not a theory question, and neither paper takes it up.

‘The Golden Rule: A Naturalistic Perspective’ (Apr. 7, 2022)

Cofnas does the close reading Singer’s argument needs and cannot survive. Singer and de Lazari-Radek require that leading thinkers of distinct traditions independently converged on a principle similar to the principle of universal benevolence, took the point of view of the universe, and treated the Golden Rule as the essence of morality. Cofnas walks through each tradition Singer cites and shows the claim falls apart on contact.
The Hillel demolition is the cleanest move. Singer treats the stand-on-one-foot exchange as showing the Golden Rule sits at the heart of Judaism. Cofnas points to Navon’s count of 80 instances where 15 different mitzvos or states of affairs are declared equal to all the Torah. Sabbath, circumcision, tzitzis, charity, living in Israel. The “essence of Judaism is the Golden Rule” reading is a projection by Reform-influenced commentators who want Judaism to look like liberal Christianity.
The Christian section earns the thesis. Jesus does treat the Golden Rule as central, but the cosmology around it has nothing to do with impartial maximization of welfare across sentient beings. Jesus promises everlasting fire to those who offend God. He tells followers to allow themselves to be abused. He takes the perspective of a particular agent, God, not of the universe. Singer’s reading absorbs the Christian Golden Rule by stripping it of its theological context and translating it into a secular humanitarian formula. The translation costs almost everything that gives the rule force in the original.
The Confucian point lands hard. Mencius called Mozi a beast for advocating inclusive care. Confucian ethics rejects impartiality. Filial piety requires preferential love. Csikszentmihalyi’s observation that the rule was often used as a metaphor for reflexivity in action rather than as a moral imperative undercuts Singer’s reading further.
The Hindu and Buddhist sections rest on thinner textual ground, but Cofnas does enough by citing Bakker and Davis that Hindu commentaries center ahimsa rather than reciprocity, and that the Buddha’s statement appears in a single passage without framing as the essence of dharma.
The constructive move is also strong. Cofnas offers a parsimonious account of why Golden Rule-style sayings recur. Moral educators across traditions hit on the same pedagogical trick: ask the listener to imagine the situation reversed, harness empathy, encourage prosocial behavior. The convergence happens at the level of teaching technique, not at the level of perceived moral truth. This account does not require any party to track a non-natural moral fact. It requires only that humans share the empathic capacity selection installed in them and that teachers across cultures notice that capacity can be tapped.
The Mozi and Anglican utilitarian section extends past pure debunking. Cofnas concedes both groups argued for impartial morality among humans. He shows both did so on theological premises. Mozi argues from Heaven’s perspective. Berkeley and Gay argue from God’s perspective. Strip God or Heaven from those arguments and impartiality goes with it. The Anglicans were clear that moral obligation requires God and God’s sanctions. Substituting “the universe” for “God” is not translation. It replaces one premise with a different premise that does not do the same work.
A few places I would push back or extend.
First, Cofnas’s empathy-pedagogy explanation is plausible but soft. It explains why the rule recurs as a family resemblance. It does not yet explain why moral educators across traditions thought empathy-induction was the right rhetorical lever. A deeper story sits underneath. Empathy works as a coordination device because it tracks a feature of social life all these traditions had to manage, the constant temptation to free-ride on cooperators. The Golden Rule reframes the cheater’s temptation by asking him to picture occupying the cooperator’s position. The rule does alliance maintenance work in each tradition, even where the surrounding theology differs.
Second, the Christian case may be more interesting than Cofnas’s debunking allows. He treats Christianity as the outlier where the Golden Rule is the cornerstone. The question Singer should ask is why Christianity made it the cornerstone. The answer might lie in the missionary requirements of a faith trying to bind together populations across ethnic lines. The Golden Rule does in Christianity what filial piety does in Confucianism and what ahimsa does in Hinduism. It is the coalition-defining moral signature. Each signature suits the social formation it serves. None reflects a universal truth. Each reflects a particular alliance structure.
Third, Cofnas’s debunking is incomplete. He notes that debunking does not show a belief is false, only that the belief is unjustified once we see its causal origin. The Singer move he targets was always a meta-ethical maneuver: use cross-cultural convergence as evidence of non-natural moral truth. Knock out the convergence and you have not refuted non-natural moral realism. You have removed one route into it. Realists can still claim the Golden Rule represents a non-natural truth. They cannot use Sidgwick’s three-part test to anchor that claim.
Fourth, the paper rests on a quiet point worth pulling out. Singer needs Christianity’s universalism to stand in for human universalism so he can recruit Christianity to the secular utilitarian project while shedding the supernatural baggage. The same move the Anglican utilitarians made. Take Christian moral architecture, swap out God, keep the impartiality. Cofnas shows the architecture cannot support the impartiality once God is removed. This matters for understanding contemporary secular liberalism, which runs on Christian moral software while denying the operating system underneath.
The paper rewards careful reading of texts in their tradition rather than the synoptic move that flattens them into a single message. The cost is that it leaves the harder constructive question open. If not the Golden Rule, what explains the recurrence of similar prosocial precepts across cultures, and what should we conclude about the standing of those precepts? Cofnas points toward an answer and does not develop it far. The empathy-pedagogy story is the start of an account. The alliance-maintenance story is what would finish it.

‘How Gene-culture Coevolution Can–But Probably Did Not Track Mind-Independent Moral Truth’ (Aug. 25, 2022)

The paper does something clever and something self-defeating at the same time.
The clever move pushes the debunking argument back a step. If Boehm is right that moral psychology evolved by social selection, with humans consciously designing rules and enforcing them, that opens the door to realist social selection. Maybe the rule-makers grasped moral truth via reason and enforced rules accordingly, so the resulting selection pressures tracked truth. Cofnas closes that door by arguing the rule-makers were themselves driven by fitness-tracking impulses, the desire to avoid domination, the desire to eat, the desire for cooperation that aids survival. Reason served those impulses rather than apprehending moral facts. So the process tracked fitness all the way down.
The self-defeating part: Cofnas relies on reason throughout to make this metaethical case. His reasoning faculty came from natural selection just like everyone else’s. If natural selection produces faculties that track fitness rather than truth, why trust his metaethical conclusion? He needs reason to track truth when he uses it to debunk moral realism, but not when our ancestors used it to design rules. The asymmetry needs argument. He gestures at Street’s “starting fund” point, that reason elaborates on initial evaluative judgments rather than escaping them, but this cuts against him. If true, his own argument elaborates on his starting fund of antirealist priors.
The “implausible coincidence” argument cuts both ways. Cofnas assumes moral truth and fitness are separable, so alignment between evolved drives and moral facts must be coincidence. Many realist traditions reject this separation. Aristotelians, Thomists, eudaimonists, and natural-law theorists hold that moral truth tracks human flourishing because flourishing is the moral fact, or close to it. The desire to avoid domination might track moral truth because unjust domination is wrong. The drive and the truth are not separate variables that happen to line up.
The Boehm thesis carries a lot of weight here. Cofnas acknowledges in footnote 2 that the strict egalitarianism of nomadic foragers is contested by Wengrow, Graeber, Singh, and Glowacki. He waves this away by saying initial conditions matter most. But if the anthropological foundation wobbles, the whole structure wobbles. The argument requires not just that early humans designed rules but that they designed them in patterns Cofnas can debunk. If foragers swung between hierarchical and egalitarian arrangements seasonally, the simple coalition-of-subordinates-against-alpha story loses force.
His treatment of liberal convergence applies coalition logic, well enough as far as it goes. Majorities reassert interests against modern alphas, so the spread of liberal values reflects power shifts rather than moral discovery. But many liberal commitments cut against majority interest. Free speech for unpopular minorities, due process for the accused, property rights against redistribution. If liberalism tracks majority power, why these counter-majoritarian features? The story works for some liberal values and poorly for others.
Cofnas runs an Alliance-Theory-adjacent argument without naming it. Moral codes track coalition interests rather than mind-independent truth. My four diagnostic questions cut deeper than Cofnas does, because they apply to him too. Who does Cofnas rely on for status, income, protection? Oxford philosophy at the time he wrote this. Who must he attract or retain as allies? Naturalist philosophers, evolutionary psychologists, the secular academic establishment that publishes EDAs. What beliefs and signals mark coalition membership? Skepticism toward moral realism, sophistication about gene-culture coevolution, willingness to debunk. What might he give up if he changed position? A lot, as his subsequent trajectory after Cambridge demonstrated. His debunking project has its own coalition function.
Two more points.
The FitzPatrick objection deserves more weight than Cofnas gives it. He claims his version differs because both sides agree on the social-selection story and disagree only about motivation. But disagreement about motivation is the whole game. Whether you describe early humans as satisfying primate desires or responding to perceived goods loads the question. The debunker frames the same behavior in fitness language and calls victory.
Second, the paper does not address why social selection produced workable societies if humans had no reliable access to anything beyond fitness drives. Pure fitness-maximization should produce something closer to chimp tyranny rebooted. That nomadic-forager arrangements look broadly defensible by independent moral standards needs explanation. “Coincidence” does too much work.

‘Still No Evidence for a Jewish Group Evolutionary Strategy’ (Jan. 6, 2023)

This third paper is the strongest of the trilogy, and the strongest because Cofnas finally drops the pretense that he’s just doing neutral science.
The interview material on MacDonald is the centerpiece. MacDonald agrees that “there are no good Jews” is “a good rule of thumb.” He agrees that the proper relation between Jews and Whites is one of service. He has “mostly positive things to say” about Hitler and the Third Reich, declines to mention the Nazi treatment of Jews when given the chance, and has gotten “more and more open” to Holocaust denial. He endorses David Duke for office. He publishes pieces in his journal calling Jews “Jugly” and arguing they “oppose beauty.” Once you have this on the record, the question of whether MacDonald is doing scholarship or coalition-tending answers itself. The 1998 trilogy was always coalition-tending. The interview just shows the coalition without the scholarly drag.
The Sweden empirics are the most underrated contribution. Cofnas runs the numbers. Across 52 countries, controlling for GDP per capita, the percentage of the population that is Jewish has no statistically significant relationship with willingness to fight for one’s country. Across 18 top migrant-destination countries, the percentage Jewish has no statistically significant relationship with attitudes toward immigrants. The Foa, Romero-Vidal, and Klassen graph showing liberal value rise concentrated in rich democracies while the rest of the world flatlines does more damage to MacDonald than any sourcing critique could. If Jews caused liberalism, you would see Jewish presence correlate with liberalism cross-nationally. You don’t. The correlation is with wealth.
The Margherita Sarfatti material is the best single section in any of the three papers. It’s also the section that most undermines MacDonald on his own terms. Italian fascism, one of the two successful fascist movements of the twentieth century, was substantially built by a Jewish woman who was Mussolini’s mentor, mistress, biographer, ghostwriter, and chief propagandist. Italian Jews supported fascism in proportions far above their population share. Mussolini turned anti-Semitic in 1936 not because the Jews had done anything wrong but because alliance with Hitler required it, and even his Jew-baiter-in-chief Farinacci could only manage the indictment that Italian Jews paid their taxes, obeyed the laws, fought in the war, but maybe didn’t love the regime enough. The same pattern shows up in the German racial hygiene movement before the Nazis, where Jewish geneticists like Weinberg and Goldschmidt were prominent supporters of eugenics until the anti-Semitic wing won out and pushed them away. Hitler told Rauschning that Jews would have flocked to his movement if he’d merely held out a finger. The historical record is clear. Jews participate heavily in nationalist movements when admitted on equal terms, and get expelled when the movements turn anti-Semitic, after which they are blamed for not participating. This is exactly the unfalsifiable structure Cofnas identified in 2018 and now has the historical evidence to crush.
The PCIN report exchange is devastating in a small way. MacDonald cites page 42. The page doesn’t say what he claims. Cofnas emails him in 2016. MacDonald says oh, I meant page 32. Page 32 doesn’t say it either. MacDonald says oh, pages 107 and 108. Those don’t say it either. They say the quota system was being bypassed because legislators couldn’t achieve their goals through it. They don’t say anything about the desirability of changing the racial balance. This is the same Hugh Davis Graham misuse pattern from the 2021 paper. It’s the same Sanford pattern from the 2018 paper. Three papers, three identical sourcing collapses. The trilogy is a sourcing-failure trilogy.
Cofnas’s positive theory of anti-Semitism in the closing section is the weakest part. Religious origins, market-dominant minority resentment, political scapegoating, and bad-is-stronger-than-good asymmetry are all real forces but they are stitched together as a list rather than a model. The Chua framework on market-dominant minorities does most of the work, and it would have been stronger to lead with that and treat the religious overlay as one factor among many rather than building it up across pages of Eusebius and Augustine quotation. The Mark Twain reference is also slightly evasive given that Twain’s “Concerning the Jews” is more ambiguous than Cofnas implies. But the basic structural point holds. Anti-Semitism shares the form of resentment directed at Overseas Chinese, Indians in East Africa, Lebanese in West Africa, Whites in Zimbabwe, and Tutsis in Rwanda. The form is older than the content. MacDonald treats the content as proof of a Jewish-specific evolutionary strategy. The form being shared across all market-dominant minorities falsifies that.
The Italian fascism section actually points toward a richer analysis than Cofnas draws. Sarfatti, Pontremoli, Jarach, Jona, the elder Sarfattis. These are not anomalies. They are predicted by Pinsof’s Alliance Theory once you drop the essentialism. Italian Jews in 1919 had a coalition structure that aligned with Italian nationalism. Italian nationalism had a coalition structure that initially welcomed Jews. The two coalitions cooperated. When Mussolini’s coalition needs shifted, the alliance broke. None of this requires a Jewish group evolutionary strategy. It requires only the ordinary logic of which sub-coalitions ally with which other sub-coalitions under which conditions. The same analysis explains why Jewish neoconservatives ally with evangelical Christians on Israel today, why Soviet Jewish refuseniks became American Republicans, why Orthodox Jews now vote Republican while Reform Jews vote Democrat. None of this is “the Jews” doing anything. It’s particular Jewish coalitions making particular alliances based on particular interests, the same way every other sub-population does.
What Cofnas still does not name, and what makes the trilogy feel slightly incomplete even at its best, is that MacDonald’s project is not failing as scholarship by accident. It is succeeding as something else. The 1998 trilogy is a coalition-formation document. It provides a moral vocabulary that explains every grievance the alt-right has by attributing it to a unified, essentialist, multigenerational Jewish agency. The book’s persistent appeal among readers who have access to Cofnas’s demolition is not despite its scholarly failure but because the scholarly failure is the price of the moral utility. A theory that could be falsified by Marcuse advocating Arab return to Israel, or by Sarfatti building Italian fascism, or by Reform Judaism actively campaigning for racial diversification of the Jewish community, would not perform the coalition function. The book has to be unfalsifiable to do what its readers want it to do. Cofnas circles this point throughout three papers but doesn’t quite say it. Saying it would require admitting that the same dynamic operates on his own side, since “race realism” performs an analogous coalition function for an analogous readership.
The retraction story is the small ugly subplot. Philosophia published MacDonald’s reply, the associate editor resigned, Weinberg agitated, Springer retracted, Kasher apologized and lost his job. Cofnas opposed the retraction publicly and was right to. Suppression of MacDonald creates the martyr halo Nietzsche described, and the martyr halo is worth more to MacDonald’s readership than the paper itself ever was. Cofnas’s instinct here is exactly correct. The way to defeat coalition-justifying pseudo-scholarship is to demolish it in the open, which is what these three papers do.
The framing of “two narratives equally false” was always a coalition position rather than a neutral one, and Cofnas’s later career makes which coalition he was joining unmistakable. Read the papers for what they accomplish. They demolish MacDonald. They do not, and were never going to, demolish the coalition structure that produced MacDonald and that will produce his successors.

‘Heterodox Academy: A Good Idea Gone Awry? | Nathan Cofnas’ (Mar. 7, 2023)

Heterodox Academy accepted the coalition it needed to attract. Haidt and Tomasi recruited respectable academics who depend on Harvard, Yale, NYU, and the broader prestige system for status, income, and protection. That coalition cannot survive a frank race-differences argument, so Heterodox Academy never had one. The drift toward respectability was not a betrayal of the founding mission. It was the founding mission, visible in the original membership list and in who they let speak from year one.
Who does Heterodox Academy rely on for status, income, protection? Mainstream academic peers and the donors who fund credentialed centrism. Who must they retain as allies? The credentialed center-left who give them legitimacy outside the right-wing media ecosystem. What beliefs and signals mark coalition membership? Quoting Mill, condemning Trump, treating McWhorter as the upper bound of permissible race talk, treating Nyborg as beyond the pale. What would they give up if they crossed the line? Everything that distinguishes Heterodox Academy from American Renaissance, which is to say everything that makes them legible to their funders and peers.
So Cofnas’s “they failed” reads better as “they did what their coalition required.” Failure is a moral verdict. The behavioral pattern is not a mystery once you ask who they needed to please.
The weaker move is his empirical confidence. Cofnas treats the genetic explanation of group disparities as settled and treats hesitation as either dishonesty or stupidity. He says “no one who understands race differences is woke” and that “knowledge of race differences cures wokeism.” This has the shape of religious testimony. Knowledge as conversion. The honest position is that the evidence is contested, the heritability-of-group-differences question runs into hard methodological problems even among scholars who accept individual heritability, and people most certain on either side tend to be those whose coalition rewards certainty. Cofnas notices this pattern for everyone except himself.
His strategic recommendation has the same coalition structure his diagnosis exposes. He wants dissident faculty to ally with DeSantis-style political pressure from outside, plus internal pushback against activist hiring, plus a public campaign on race differences. The proposal might work if the coalition holds. It probably will not hold, because Republican governors operate on election cycles, dissident faculty operate on tenure horizons, and race-realist scholars operate on the need to be heard. Each group needs something from the others that the others cannot reliably deliver.
Heterodox Academy is a procedural project that pretended to be neutral about substance. Procedures cannot stay neutral when the substance is contested. They get captured by whoever pays for them and whoever grants them legitimacy. Haidt knew this in theory and acted as if he could escape it in practice. He could not. Tomasi inherits the same trap. His reply to Cofnas, that Heterodox Academy “is not an anti-anything organization,” is the trap announcing itself. An organization that stands against nothing stands for nothing the orthodoxy can be made to fear.
What is left for someone who wants heterodoxy? Probably not another Heterodox Academy. Probably not a new university either, since the credentialing pipeline and donor base pull any new institution toward the same equilibrium within a decade. University of Austin will face the same forces in five years that Heterodox Academy faced in three. The realistic options are smaller. Protected fellowships funded by donors who do not need elite approval. Individual scholars who write for audiences outside academia. The slow accumulation of work that becomes citable when the orthodoxy weakens.
Cofnas is doing the third one.

Matt (History Speaks): ‘A Chat with Nathan Cofnas’ (Jun. 6, 2023)

Apply the four questions to Matt. Status, income, protection: a PhD candidacy at LSE in international history, the YouTube channel as a side project, the eventual historian career. Allies: serious historians who engage rather than dismiss, free speech academics, the small subset of internet history audiences willing to watch debates with Holocaust deniers. Membership signals: defending free speech for hereditarians while disagreeing with the substance, debating deniers directly, identifying MacDonald’s bad-faith moves, his Coptic background as a live counterexample to simple hereditarian inference. What he gives up if he changes position: the protective cover that most academics use. He has chosen the costly route of engaging instead of denouncing. The position is structurally similar to Cofnas’s, which is why the conversation works as well as it does.
Matt does what Heterodox Academy claims to do better than Heterodox Academy does it. He is not academy-protected. He is a doctoral student running a YouTube channel. Notice the implication for any model of intellectual courage that requires institutional cover. The cover is not the source. The willingness is.
The hereditarian exchange is the most revealing in the interview because Cofnas retreats to a weaker claim than the one he carries elsewhere. Matt presses with the Coptic example. Christians from the Levant and Egypt, descended from civilizations that built the foundations of the West, score low on cognitive tests today. Cofnas’s reply is that we lack reliable data on those populations because we cannot measure their potential under controlled conditions. This is a retreat. The strong hereditarian claim says current scores reflect genetic potential after corrections for environment. The retreat says we cannot make that judgment without cross-population adoption studies. The retreat is more defensible. It is also weaker than the claim he uses to drive his political argument. Matt does not press the gap. The gap stays.
The wokeism exchange is where Matt hits the keystone directly. He grants the empirical claims for the sake of argument and asks why the political conclusions follow. If hereditarian differences exist and we cannot change them, why must we make incendiary public arguments? Why not the quieter conservative position that says differences exist, we do not know how to change them, so we maintain meritocratic standards? Cofnas’s reply is that wokism logically follows from the equality thesis, so the equality thesis must be defeated to defeat wokism. Matt rejects the logic. He is right to reject it. Wokism does not logically follow from the equality thesis any more than Catholicism logically follows from Genesis. Movements survive partial concessions on their empirical premises. They adapt. The keystone framework Cofnas keeps reaching for has the shape of an essence, which is what Turner would call the error. The work of a movement is not done by any single claim. It is done by the institutions, signals, and incentives that the movement maintains around the claim. Refute the claim and the institutions adjust.
The MacDonald material is where Cofnas operates at his most rigorous. The argument has been worked through. Jewish overrepresentation in leftist movements is matched by Jewish overrepresentation in basically every intellectual movement, including fascism through Margherita Sarfatti, including libertarianism through Mises and Rand and Friedman and Rothbard. The default hypothesis of average IQ around 110 plus urban concentration plus the pattern of Diaspora intellectual involvement explains the data better than the group-evolutionary-strategy theory. The intermarriage rate of eighty percent among secular American Jews destroys the racial-purity claim that MacDonald said was the goal of the strategy. MacDonald’s response, that intermarriage helps strategically through cases like Kushner-Trump, Cofnas correctly identifies as the kind of post-hoc rescue that Holocaust deniers use to preserve their theory against contrary evidence. The comparison to Holocaust denial is sharp because both projects share the same epistemic structure: a predetermined conclusion defended by finding logically possible explanations that lack evidentiary support.
The Holocaust denial discussion is where Cofnas and Matt agree most clearly on method. Engagement is better than censorship, both because censorship drives belief underground where it metastasizes, and because direct engagement reaches the small percentage who can be moved. Matt has the receipts. Two former TRS subscribers deprogrammed. The framework here is the same one Cofnas uses against Heterodox Academy. The orthodox refusal to engage has produced the radicalization the orthodoxy claims to prevent. Engagement is the intervention.
The conservative-movement-as-anti-intellectual complaint at the end is the seed of what becomes Cofnas’s theory of right-wing stupidity. He cannot publish in legacy conservative outlets despite four hundred thousand paper downloads. The outlets reject him on the title. They want Chuck Schumer hypocrisy pieces and Elizabeth Warren ancestry jokes. He attributes this to lack of intellectual seriousness. The deeper reading he does not give is that his hereditarian message threatens conservative pieties as much as progressive ones. Conservative meritocratic individualism has its own version of the equality thesis. Anyone can succeed if he works hard and lives right. Hereditarianism complicates that story. The legacy conservative outlets reject Cofnas for the same coalition reasons the philosophy mainstream rejects him. He sees one and not the other.
The metaethics contradiction is on display the way it is on display in every Cofnas interview. He spends fifteen minutes arguing that natural selection has not equipped us with moral intuitions that track moral truth. The debunking arguments work, he says. Moral realism makes no sense. Then he spends the rest of the interview making confident moral claims about wokism, hereditarianism, intellectual honesty, the right to free inquiry, and the moral failings of conservative grifters. The two halves cannot both stand. Either the claims have purchase, in which case the debunking was overstated, or they do not, in which case the rest of the interview is empty. Self-deception is the cheapest hypothesis. The contradiction would dissolve the project, so the mind protects the project from recognition. Trivers wrote the model. Cofnas inhabits it.
The Becker layer runs through both men. Cofnas builds meaning through hereditarian truth-telling as the master key to the political pathology. He is the man who sees what others cannot and pays the price for seeing. Matt builds meaning through engaging the deniers directly. He is the historian who refuses the protective ritual of denouncement and chooses the costly ritual of debate. Both hero systems require the existence of an enemy worth fighting. Both produce real intellectual work. Both supply the practitioner with a story about why his particular labor justifies his particular life. Neither story collapses on contact with the other. They coexist in the conversation because both men recognize the other as doing structurally similar work.

‘Talkline With Zev Brenner With Nathan Cofnas on the vanishing Liberal Jew’ (Sep. 20, 2023)

Zev Brenner is a Jewish broadcaster doing what Jewish broadcasters do, which is spending the last twenty minutes of any interview with an unaffiliated Jew gently trying to bring him home. The structure is familiar from a thousand similar conversations. The interesting thing is what the conversation accidentally reveals about Cofnas’s position before any of his other coalition commitments crystallized into the framework you have now seen across seven later interviews.
Apply my four questions. In September 2023 Cofnas had not yet started his Substack. The Cambridge controversy had not erupted. The lawsuit did not exist. The hereditarian-revolution essays had not been written. He was a Cambridge research fellow doing philosophy of biology with side commentary on Jewish demographics in the small magazine The Critic. His coalition position was much less load-bearing than it would soon become. He depended on Cambridge institutional standing and on whatever readership his occasional essays attracted. He needed to attract neither the hereditarian coalition nor the Free Speech Union pool nor the heterodox-academic-podcast circuit, because none of those had become relevant to his career yet. The interview is conducted before the framework hardened, and it shows.
The Twilight of the Liberal Jew thesis is straightforward demographic projection. Liberal Jews intermarry at over seventy percent. Their identity dilutes within two generations. Orthodox Jews reproduce at replacement-plus rates. The demographic transition is mathematical. By the second half of the twenty-first century the visible American Jewish community will be predominantly Orthodox, Trump-voting, and politically conservative, and the secular liberal Jewish public figure will be a historical artifact. The thesis is correct as far as it goes. The demographic arithmetic is not in dispute among people who study Jewish demographics. Cofnas is reporting what Steven M. Cohen and other Jewish demographers have been documenting for decades.
What is more interesting than the thesis is what Cofnas does with it and how Brenner responds. Brenner asks the obvious follow-up question. Are you Orthodox yourself. The answer is no. Brenner asks where Cofnas places himself. The answer is politically right, attended yeshiva briefly, not part of the Jewish community now. Brenner asks whether there is a Chabad in Cambridge. Cofnas does not know. Brenner asks whether Cofnas is married. Cofnas mentions, with what reads as careful neutrality, that he has married a Korean woman. The interview ends with Brenner offering Shabbat dinner if Cofnas is ever in town and Cofnas saying we will see, possibly something will change in five or ten years.
The personal disclosures matter for what they show about the framework that Cofnas would later assemble. Cofnas presents himself variously as a Nietzschean transhumanist, a credentialed defender of free inquiry, a hereditarian strategist, a careful philosopher of biology, and a critic of MacDonald who happens to be Jewish. The Brenner interview captures the substrate beneath all of those self-presentations. He is a partly-deracinated American Jew who attended yeshiva, retains intellectual interest in Jewish demographic questions, has married out, and has chosen a personal trajectory that exemplifies the very pattern his Twilight essay describes as a tragedy. He is, in Brenner’s gentle but accurate framing, writing about his own community’s disappearance while contributing to the disappearance.
Cofnas says “I think it is really a tragedy that Jews were never able to find a way to express their cultural identity outside of the orthodoxy in a way that was sustainable.” Brenner asks whether Cofnas wants to be part of maintaining that tradition. Cofnas says it is a bit late. The exchange is precisely the contradiction that animates the project. Cofnas mourns the loss of a sustainable secular Jewish identity while recognizing that he has personally placed himself outside the only configuration that has proven sustainable. The mourning is real. The choice is also real. Both are present and neither cancels the other.
Cofnas’s framework requires that he hold positions that do not cohere, and the Pinsof framework predicts this will happen because coalition position-taking does not require internal consistency. The Brenner interview shows the personal-life version of the same structure. Cofnas mourns assimilation while assimilating. He critiques wokeism for failing to ground its moral authority while operating from metaethical commitments that undermine his own moral authority to critique. He calls for hereditarian institutional capture while building a career path within the institutions he says are unfit. He defends Jewish over-representation in elite positions against MacDonald-style anti-Semitic framings while documenting Jewish demographic decline in ways that would be congenial to those framings if read by hostile audiences. The structure is consistent. He holds positions that work for different coalitions, and the positions do not need to cohere because no single audience reads all of them together.
Brenner does something interesting in the second half of the interview that no other interlocutor in later interviews attempts. He treats Cofnas as a person rather than as a public position-taker. He asks about Cofnas’s family. He extends an invitation to Shabbat dinner. He observes, accurately, that Cofnas has a soul that he can feel. The framing is mildly embarrassing in the way that Jewish broadcasters speaking to lapsed Jews are mildly embarrassing, and Cofnas handles it with appropriate awkwardness. But the framing surfaces something the later interviews never get near. Cofnas has a personal life that has shaped what he writes. The Twilight essay reads differently when you know the author married out. The hereditarian-revolution essays read differently when you know the author identifies politically with the right but has chosen a partner from a population that the populist hereditarian coalition would treat as a marker of racial dilution. The Nietzschean self-identification reads differently when you know the author attended yeshiva briefly and writes essays about Jewish demographic decline.
Cofnas operates across multiple coalitions whose memberships do not overlap, and his personal life is structured the same way. He is a Jewish-by-ancestry intellectual who has personally exited the Jewish community, married out, and chosen a partner from outside his ancestral group. He writes for the hereditarian coalition while exemplifying the deracinated cosmopolitan pattern that the harder-edged hereditarians treat as civilizational decline. He writes for the heterodox-academic free-speech coalition while sitting at Cambridge in the kind of credentialed position that the populist right treats as evidence of institutional capture. He writes for Jewish demographic readers while disclosing that he himself is the disappearance he describes. None of this is hypocritical in a strong sense. It is what coalition position-taking looks like when the position-taker has commitments that pull in different directions. The personal life and the intellectual project run on the same logic.
Becker’s hero system frame applies here. The Twilight essay is not just demographic analysis. It is also a kind of self-aware participation in a tradition Cofnas has left. By writing about Jewish demographic decline he locates himself within the Jewish intellectual tradition of mourning Jewish demographic decline, which is a recognizable subgenre going back at least to the late nineteenth century. He cannot be in the community in the way Brenner means. He can be in the conversation about the community. The Twilight essay performs that being-in-the-conversation. The performance is not insincere but it is also not the same thing as the affiliation Brenner is gently inviting him to.
Trivers on self-deception applies to the gap between the demographic prediction and the personal trajectory. Cofnas writes the essay as if from a position of demographic and political analysis, which is the framing the essay uses. But the essay also reads as something the author needs to write because writing it lets him hold his ancestral identity and his exit from it in the same intellectual space. He can be the analyst of Jewish decline rather than an instance of it. The framing is not a lie. It is the kind of repositioning that lets a person live with a choice that has costs the person has not fully metabolized. Brenner is good enough at his job to feel this and gentle enough not to press it too hard.
The political-prediction component of the essay deserves separate attention because it has aged interestingly. Cofnas predicts that the visible American Jewish community will become predominantly Orthodox and politically conservative within a generation or two, and that the liberal Jewish public figure will become a historical artifact. Two and a half years on from this interview, the prediction is partly tracking and partly not. The Orthodox demographic growth is real. The political shift among non-Orthodox Jews has been slower than Cofnas’s framing suggested. The Israel-related events of October 2023 and after have produced complicated political realignments that do not map cleanly onto the Orthodox-versus-secular axis. Many secular Jews who were Democrats in 2023 have become more politically heterodox, and many Orthodox Jews remain on the right but with significant variation. The simple Orthodox-conservative versus secular-liberal binary that Cofnas’s essay assumed has proven less stable than he projected. This does not invalidate the demographic arithmetic. It does suggest that the political trajectory is shaped by factors the demographic frame underweights.
The interview also shows Cofnas’s ambivalence about his own framework more clearly than any of the seven later interviews. When Brenner asks whether Cofnas wants continuity for himself, Cofnas says it is a bit late, and the framing carries something that reads as regret. The regret is not visible in the later interviews because the later interviews are conducted within frames that do not invite it. Razib does not invite it because Razib is engaged with intellectual content rather than personal life. Boyce does not invite it because Boyce is interested in the framework rather than the man. Green does not invite it because Green is contesting epistemology. Brenner is the only interlocutor who pursues the personal question to the point where the regret surfaces, and it surfaces clearly.
The four questions, applied to Cofnas’s personal life: Whose status, income, and protection does he depend on. Cambridge institutional standing, the small philosophy-of-biology coalition, his Korean wife. Whom must he attract or retain as allies. Other heterodox academics, free-speech advocates, hereditarian-curious readers, and within his domestic life, his wife and her cultural community. What beliefs and signals mark coalition membership. Scientific framing of contested empirical questions, defense of free inquiry, careful prose, intermarried-secular-Jewish intellectual identity that does not require Orthodox affiliation. What would he have to give up if he changed his public positions. The career, the heterodox-podcast circuit, the Substack readership, and within his personal life, the framing of his own life trajectory as analytically chosen rather than the product of coalition pressures the framing was selected to manage.
The Brenner interview does not develop any of these themes because Brenner is not running the diagnostic. He is running a different operation, which is the standard Jewish-broadcaster operation of seeing whether a returnable Jew might be returned. He concludes that the patient is not yet ready but might be ready in five or ten years. Cofnas demurs gently and the interview ends. The exchange is decorous and slightly sad and tells you something the later interviews do not.
Brenner ended his interview by saying it should be a happy show and a good year, the standard Jewish broadcaster sign-off in late September approaching the High Holy Days. The timing is appropriate because the High Holy Days are the season of accounting, and the interview is itself an accounting that Cofnas does not quite make but does not refuse either. He says we will see, and then five years pass, and the framework has hardened, and the interviews together show what the hardening has accomplished and what it has cost.

‘Philosopher Nathan Cofnas on DEI & the Middle East Conflict’ (Oct. 27, 2023)

Cofnas occupies a position few people occupy: hereditarian on race, anti-MacDonald on Jews, pro-Israel on the post-October-7 question. The position is structurally rare and structurally fragile. Each commitment costs him a different coalition. The hereditarianism costs him the philosophy mainstream. The anti-MacDonald work costs him the alt-right intellectuals who might otherwise be his most enthusiastic readers. The defense of Israel costs him the part of the dissident scene that has gone hard against Israel since Gaza started burning. He keeps all three because dropping any one would collapse the niche, and the niche is the asset.
I press him on the Jewish question several times. Each time he runs the same move. He grants the surface point, denies the structural one, and shifts to a demographic observation about declining Jewish numbers. Yes wasps had self-hatred too. Yes Jewish liberals jumped on the diversity bandwagon. Yes there are Jews who are part of the problem. But the deeper claim that Jews orchestrated the postwar liberal turn fails empirically. Countries without Jews behaved similarly. MacDonald is wrong. This is a coalition signal as much as an empirical claim. He needs the alt-right reader to understand he will not go where MacDonald goes, and he needs the philosophy reader to understand he is not doing tribal apologetics.
The October 7 timing puts his framework on a stress test. His general theory says wokeism follows from the equality thesis applied to persistent disparities. The progressive coalition then hunts for hidden racism to explain what the equality thesis cannot account for. After October 7 the same logic runs on Israel. Israel is White-coded, privileged, colonial; therefore Israeli grievance is White grievance and Palestinian violence is the reaction of the oppressed. Cofnas has been predicting this kind of move for years. The interview shows him watching his framework run on live data and calling the result.
His one moment of surprise is worth pausing on. He says he was surprised by Democrat establishment support for Israel. He attributes the support to octogenarian Boomer holdovers and predicts the Zoomer left will not feel the Holocaust association. The prediction tracks how coalition replacement works. Older members hold older signals. New members carry the new signals. Sympathy with Jewish suffering as more weighty than other White suffering is a Boomer signal, not a structural one.
The BDS exchange is the cleanest moment of coalition language replacing argument. I catch him in what looks like bad faith. Pro-Israel voices ask Palestinians to develop nonviolent means, Palestinians develop BDS, pro-Israel voices oppose BDS. Cofnas’s reply is that BDS is “an active war,” which dissolves the deliberative frame entirely. War is the failure of deliberation, not a kind of deliberation. Once he classifies BDS that way he does not have to engage the Palestinian case for nonviolent pressure. The move is what he accuses progressive academics of doing when they classify hereditarian work as harmful and therefore outside discourse.
Whatever the truth about heritability, an institution that claims to defend heterodoxy while excluding the heterodox positions that most threaten the orthodoxy is doing coalition work, not heterodoxy work.
Turner on essentialism applies here. Cofnas keeps reaching for the equality thesis as a single load-bearing claim that, if dislodged, brings the whole structure down. The claim has the shape of an essence. Wokeism is what you get when you take the equality thesis seriously. Refute the thesis and you cure the disease. The framework is too tidy. Turner’s point is that movements sustain themselves through unresolved tensions, not through any single keystone. Wokeism will survive partial concessions on heritability the way Catholicism survived partial concessions on Genesis. The believers who care about the science adjust. The institution moves on. Cofnas’s strategic recommendation, that hereditarian truth-telling will dissolve the wokeist project, depends on the equality thesis carrying the whole load. It might be one stone among many.
The Trivers move is the metaethics contradiction. Cofnas says natural selection has not equipped us with moral intuitions that track moral truth. He then spends the rest of the conversation making confident moral claims. Wokeism is wrong. Israel has a strong moral claim. Palestinians have a weaker one. Vegetarian propaganda is harmful. The claims sit on the same evolutionary substrate his metaethics was supposed to debunk. He cannot have it both ways. Either moral claims have purchase, and the debunking was overstated, or they do not, and the rest of the interview is empty rhetoric. Self-deception is the cheapest hypothesis. The contradiction would dissolve the project, so the mind protects the project from recognition.
Cofnas says the liberal Jew will fade because the numbers are dropping, the heritage is mixing, and the conservative Jew is rising. He names Ben Shapiro and Libs of TikTok as the prominent young Jews and points to his own essay, “Twilight of the Liberal Jew.” The observation is sharp. It also shows what kind of meaning his framework supplies him. The hereditarian project, on his view, is a Boomer-replacement story. The old gatekeepers retire, the young inherit, the numbers shift, and the truth wins on a generational timeline. This is a hero system structured as historical inevitability. The denial of death runs through the framing. Cofnas can be the man who saw it coming, paid the costs early, and appears at the end of the story as having been right.
His own analogy at the close gives the game away. He cites the priestly initiation in Leviticus where the new priests are shaved and waved and made to look ridiculous, and he reads the ritual as preparation for leadership through accepted humiliation. He has read his own situation correctly. The dissident position requires absorbing reputational damage in advance, treating the damage as the price of standing somewhere most people will not stand. What he does not say, and probably cannot say, is that the priestly initiation works only because the priesthood that follows is real. If the office is empty, the humiliation is just humiliation.

‘The Problem of Intelligence – IQ and Its detractors (Guest: Nathan Cofnas)’ (Dec. 4, 2023)

Two Swedish dissident-adjacent podcasters who agree with most of what Cofnas says before he says it. Carl describes himself as a thoughtful nationalist-curious intellectual. Yan calls himself a vulgar Marxist. They open with a Jewish joke that tests whether Cofnas is comfortable in this space, and Cofnas is. The interview that follows is structured around the hosts’ concerns more than his. They want to know what comes after wokeism. He does not have an answer. The mismatch is the most revealing thing in the conversation.
Apply the four questions to the hosts. Status, income, protection: a small podcast, the Swedish dissident-intellectual scene, presumably day jobs that pay the bills. Allies to attract or retain: race-curious intellectuals, Cofnas-style heretics, the broader European nationalist-adjacent podcast network. Membership signals: the Jewish opening joke, the references to Benedict Anderson and Anthony D Smith on nationalism, the Coca-Cola hilltop commercial as a cultural artifact for the dying universalist project, the Strangelove gag at the end about three Jews and one Nazi building the bomb. What they give up if they change position: their show.
The hosts are using Cofnas as a foil for a larger project. They want to think through European decline. The Coca-Cola commercial is their key text. The dream of liberal universalism is dying, in their reading, and something must replace it. They press Cofnas toward three possible replacements in turn. Anthony Smith’s ethnic nationalism. Singapore-style state quotas. A return to religious community. He politely declines each push. The declining is the most revealing part. He is a hereditarian truth-teller, not a nationalist or an interventionist or a religious revivalist. The truth-telling is the intervention. Anything beyond that he does not have a plan for.
Yan’s revealed-preference argument is the sharpest moment in the interview and Cofnas does not answer it. Yan says the West already practices a kind of liberal Eugenics. People abort fetuses with detected cognitive disabilities and treat the choice as automatic. Educated people select mates by credential, which is a proxy for intelligence. Universities operate as sorting and mating systems. The taboo on discussing race differences sits next to a revealed preference for selecting on traits correlated with race differences. The hypocrisy is doing real work. People claim equality while practicing selection. Cofnas glances at the argument and changes the subject. The argument has bite that he cannot use, because acknowledging it would force him to say either that the revealed preferences are good and should be made explicit, which costs him the philosophy mainstream, or that the revealed preferences are bad and should be reformed, which costs him the dissident audience. He stays quiet.
The Singapore concession is the moment Cofnas reveals he is not a libertarian on race. The hosts ask what follows from his framework. He says, “I have my own preferences I guess it would be nice if we would just treat everyone as individuals or whatever but I don’t think that’s realistic.” Then he describes Singapore approvingly. Quotas for ethnic groups in residential areas. State management of intergroup competition. Acceptance that different groups have different outcomes and that the state should arrange affairs to keep the peace given the differences. This is a substantial position. It is also a position he avoids elsewhere. The friendly interview lets it slip out.
The dysgenic fertility comment is buried in the middle of the IQ discussion and the hosts do not press him on it. Cofnas says the Flynn effect has reversed. Less intelligent people have more children. Smart people delay reproduction or skip it. Population intelligence is now declining because dysgenic pressure has become stronger than the test-sophistication trend that drove the Flynn effect upward. This is a real claim with real implications. The hereditarian who worries about declining population intelligence is structurally committed to some kind of natalist or eugenic intervention. Cofnas does not propose one. He says most people do not care about dysgenic trends any more than they care about racial composition. The observation is consistent with his general position that the political project he wants is impossible.
The religion exchange is honest. Carl raises Anthony Smith and the role of religious community in producing durable national identities. Cofnas concedes that religion did community better than secular liberalism does. He says the only way to build that kind of community now would be to invent a religion and lie to everyone about it. He cannot endorse that project. The reason he cannot endorse it sits inside his metaethics. Morality is not real. Religion is functional but false. He will not recommend a useful lie. The position is consistent with what he says elsewhere about debunking moral realism. It also leaves him with no positive program for what comes after wokeism, because every positive program requires moral commitments his metaethics cannot underwrite. The metaethics that makes him a hereditarian truth-teller also makes him incapable of building anything where the truth-telling lands.
The Korea anecdote serves as the costly-signal humor of the interview. Cofnas was deplatformed at a top Korean university, probably by a Korean academic with a UK PhD who had absorbed the Anglo-American taboos. “Too racist for Korea” as a line on the CV. The joke fits the Becker hero-system role. Cofnas is the man who saw what others could not see and paid the price for seeing. Each new institutional rejection becomes another data point in the story. The hero system requires the persecution. The persecution justifies the hero system. The Korean wife mentioned earlier in the interview functions as the counter-example that proves he is not a racial tribalist. Phrenology class as romantic origin story. The combination is the brand.
The closing Strangelove gag is coalition signal as comedy. Three Jews and one Nazi designing the bomb. The hosts joke that they are becoming more Jewish through this conversation, more dissident, more on the right tail of the bell curve, while making jokes about the master race and the Persians. This is the kind of humor only available inside the dissident-friendly podcast ecosystem. The fact that Cofnas plays along signals that he is comfortable here in a way he is not on a mainstream show. The Vera Lynn song that closes the episode, “We’ll meet again,” is the punchline. The song was Strangelove’s closing music as the bombs fell. The hosts are gesturing at end-of-civilization humor while also building their dissident network meeting by meeting.
Turner on essentialism applies as it has applied throughout. The equality thesis carries the entire weight of the framework. Cofnas treats it as the keystone whose removal collapses wokeism. The hosts do not press him on this because they agree with him. The essentialist error stays invisible in the friendly interview because the friends share it.
Cofnas concedes that evolution does not select for intelligence. Evolution rewards charisma and reproduction. Smart people have fewer children. If evolution does not select for intelligence, the long-run trajectory of intelligence is downward, not upward. The hereditarian revolution Cofnas wants is therefore not historical inevitability. It is a small window in a longer arc that bends the other way. He does not connect this dot. The framework needs the inevitability to provide meaning. The mind protects the framework from the recognition that the inevitability runs the wrong direction.

‘Why We Need to Talk about the Right’s Stupidity Problem’ (Jan. 2, 2024)

Cofnas argues that wokism follows from the equality thesis plus Christian morality, that smart people pick wokism because they follow this logic while conservatives don’t, and that the only path out is attacking the equality thesis. The piece has strengths and a load-bearing weakness.
The strength is the institutional capture argument. He correctly sees that Rufo’s philosophy story and Hanania’s law story both fail to explain why elites accepted the new framework so quickly. His point that critical theory walked through an open door fits the Alliance Theory angle reading. The elites already shared the moral premises. Marcuse and Bell gave them vocabulary.
The weakness is the prescription. Cofnas treats the equality thesis as a Big Lie that, once exposed, brings elites to the right. Apply my four questions to any elite at Harvard or Google. Who supplies their status, income, and protection? Their coalition. What signals mark coalition membership? Public alignment on race and sex. What do they stand to lose by defecting? Their entire position. Hereditarianism does not reset these forces. Plomin has spent forty years publishing careful behavioral genetics. No elite defection followed. The taboo holds because the incentives hold.
Cofnas also runs a Becker hero system without naming it. The race-realist intellectual stands as the brave truth-teller against the lie that runs the West. His own income and status now flow from this position. He lost his Cambridge fellowship over it. He writes for Aporia. The same diagnostic he applies to conservatives applies to him: he relies on a coalition, attracts allies through coalition signals, and would lose status, income, and belonging if he changed his public position.
The IQ evidence is softer than he treats it. WORDSUM is a vocabulary test. Vocabulary tracks years of schooling and verbal exposure as much as g. Liberals stay in school longer and read more credentialed prose. Kirkegaard’s 8.5-point gap among Whites by ideology partly captures education and verbal coalition signaling rather than raw cognitive ability. Cofnas notes that White race-realists score 8.5 WORDSUM points below White environmentalists. He explains this away as the wrong race-realists. A cleaner reading: WORDSUM rewards mainstream verbal conformity, and any anti-mainstream position drags the score. His own data cuts against his story.
The Turner angle is where Cofnas is most exposed. The whole essay rests on essentialist categories: intelligence as stable trait, race as stable population, conservative and liberal as stable identities. Turner’s argument against essence in social explanation lands hard here. Cofnas describes forces that exist. Smart people do cluster left in current institutions. But the explanation runs through coalition position, credential pipelines, and selection effects in elite institutions, not through the smart-people-follow-logic story Cofnas tells. Universities select for a personality type that combines high verbal ability with deference to professional consensus. That selection produces the gap he measures.
His best move is the institutional one. The right has not built serious knowledge institutions because the right does not value them enough to fund and protect them. Tucker’s 2009 CPAC line about gathering news is the cleanest evidence, and his later turn to UFOs and Obama gay-affair material confirms it. The right’s anti-intellectualism is a coalition signal, and the signal repels the people who could build the institutions.
His worst move is the strategic one. Suppose race realism became the consensus tomorrow. Disparities might still trigger demands for redress under any moral framework that treats outcomes as a public concern. The Christian moral inheritance he names is older and deeper than the equality thesis. Even hereditarians within that moral frame end up arguing for compensatory programs. The Big Lie story makes the equality thesis carry too much weight.

‘A Guide for the Hereditarian Revolution’ (Feb. 5, 2024)

The argument has a structural problem Cofnas does not see. He says elite wokism follows from the equality thesis, and the cure lies in getting elites to accept hereditarianism. He also wants to use coalition logic when convenient. The two stories pull opposite ways.
If elite belief floats free of coalition position, evidence might shift it. The Darwin analogy assumes this picture. If elite belief tracks coalition position, evidence does not move it. The same elites who might need to accept hereditarianism are the ones whose status, income, and protection rest on the equality thesis as public faith. My four questions cut here. Who does the Harvard administrator rely on for status, income, and protection? Who must he attract and retain? What signals mark membership? What does he give up if he switches? The answers explain why the strategy fails before it starts.
Cofnas’s own example refutes him. He notes that fifty years of Sowellism convinced almost no elites. He treats this as evidence that Sowell’s cultural account is too weak. The simpler reading: empirical claims about group differences, true or false, do not move elite belief in the direction Cofnas wants. Sowell offered a less taboo-violating version and got nowhere. The taboo does the work, not the empirical content. A more taboo-violating claim might face worse traction, not better.
The Darwin parallel does not hold. Darwinism slotted into a progressive story British elites already favored. It threatened ecclesiastical authority that was already eroding. It did not require the elite to lose its own coalition position. Hereditarianism contradicts the moral substrate of contemporary elite legitimacy. The two cases differ in ways that decide the question.
The argument has a striking structure: only people willing to say what Cofnas says can save civilization. Murray is timid. Sowell is too soft. Hanania too narrow. Rufo too tactical. White nationalists too low-IQ. JQ obsessives too embarrassing. Coleman Hughes too colorblind. The map narrows until only Cofnas remains. The self-deception module does its work. The world’s needs and Cofnas’s status interests align with suspicious neatness.
Apply Becker. The essay offers a hero system. Truth-tellers face the taboo and pay the price. They become Christian martyrs of the new dispensation. Cofnas knows what to do with the comparison. He cites it. The reader who accepts the framing inherits a redemptive role.
Apply Turner. Cofnas treats wokism as an essence with one root cause. Pull the root, the structure collapses. Turner’s critique of essentialism applies here. Wokism, if the term means anything, is a coalitional formation with many sources. Civil rights law, Cold War liberalism, Christian ethical inheritance, professional-class status signaling, university admissions politics, philanthropic foundation patterns, corporate HR practice, demographic change, media incentives. None reduces to belief in the equality thesis. Knock down the thesis and the coalition adjusts its signal set. The coalition does not collapse.
The empirical claim is shakier than Cofnas suggests. Polygenic scores for cognitive traits do not validate across ancestry groups. GWAS samples skew European. Cofnas’s camp has promised genetic vindication for forty years and not delivered the goods. The honest scholarly position holds that we do not know what the gene-environment partition looks like for between-group cognitive differences. Cofnas writes as if the science were settled, then complains that hereditarian scholars who hedge are timid. Their hedging might reflect calibration to the evidence rather than cowardice.
The “what comes after” section reads thin. He gestures at communities organizing along different values, minimal representation guarantees in transition, racial tribalism as a solution to collective-action problems. None gets specified. The end-state remains a placeholder.
The strongest part of the essay is the negative argument that legal-reform-only and institutional-capture-only strategies leave the underlying coalition position untouched. He is right that Hanania’s “permanent cognitive dissonance” looks unstable. He is right that Rufo’s institutional siege does not address why the institutions went left. He is wrong about what fills the gap. Coalitions shift when their cost-benefit changes. That happens through demographic change, fiscal pressure, status realignment, and the failure of coalition signals to track reality closely enough to hold credibility. Hereditarianism might eventually be one signal among many in a future elite coalition. It will not be the lever that moves the coalition.

Randy Bock, MD: ‘Controversy at @CambridgeUni: Nathan Cofnas Faces Expulsion for Challenging Ideas’ (Mar. 3, 2024)

The metaethics passage is identical in structure to the Boyce interview but more compressed. Cofnas runs the same argument. Natural selection does not track moral truth. Our moral intuitions are debunked by their evolutionary origins. He spends fifteen minutes establishing moral anti-realism and then proceeds to make confident moral claims about wokeism, free speech, and the proper organization of society. The contradiction is not addressed because Bock does not press it. The repetition across two interviews two months apart suggests this is not improvisation. Cofnas has internalized the argument as a stable component of his self-presentation, and the contradiction is also stable. The framework requires moral anti-realism for the philosophical sophistication move and moral realism for the political project. Both halves are needed and neither half can be abandoned.
The Hamas footage exchange is where Bock pushes hardest on something the framework does not cleanly handle. He cites the GoPro footage of the Hamas operative phoning his mother to celebrate killing Jews, ecstatic about the murder. Cofnas’s response is to distinguish doing something for fun from doing something during which one has fun. The Hamas operative was killing for the war, not for fun, and was experiencing pleasure incidentally. This distinction is technically correct and morally evasive. The question Bock was asking, whether the operative’s actions were morally wrong, gets bracketed. Cofnas redirects to the abstract metaethical point that some people have deviant moral intuitions and that this supports his anti-realist view. The redirection is what the framework requires. If he accepts that the murders were morally wrong in any robust sense, the metaethical anti-realism gets harder to maintain. If he accepts the metaethical anti-realism, he cannot say the murders were wrong in the sense Bock is asking about. So he splits the difference and answers neither question.
The animal cognition section is one of the more substantive passages. Cofnas correctly notes that the standard interpretations of fairness in primate experiments overstate what the data show. The capuchin grape-and-cucumber experiment does not demonstrate fairness intuition, since the same response occurs when no second monkey is present and the animal merely sees grapes. The point is well taken. But the structure of the move is worth noting. Cofnas deploys a debunking argument against animal moral cognition that uses the same form as his debunking of human moral cognition. The animals do not have what they appear to have. The humans do not have what they appear to have. In both cases the appearance of moral content is reducible to something else. The reductive move is consistent. What is not consistent is the political project that builds on the human reduction. If the same debunking applies to human moral intuitions, then the moral content of the political project Cofnas advocates is also reducible. He does not extend the debunking to his own positions. The framework requires that the debunking apply selectively.
The Nigerian and Jamaican immigrant exchange shows Cofnas at his most analytically careful but also at his most coalition-bound. Bock makes the empirical observation that Nigerian-American outcomes exceed white American outcomes on standard measures. Cofnas correctly notes that Nigerian immigrants are a highly selected sample and that comparing selected immigrants to population-representative groups distorts the inference. This is a genuine point about selection effects and Cofnas handles it well. But notice what he does next. He concedes that low-performing white populations exist and attributes their performance to genetic potential influencing culture. The framework allows variation within races to be explained by genetics, allows variation between races to be explained by genetics, and allows immigrant selection to be acknowledged as confounding the between-race comparison. The framework gives him moves for any data pattern. Whatever the data show, hereditarianism explains them. Karl Popper would call this unfalsifiable.
The Bock follow-up about whether genes matter when the same genes produce different outcomes in different settings does not get pressed. The Nigerian-American case is genuinely interesting because the same Igbo and Yoruba genes that produce middling outcomes in Nigeria produce upper-class outcomes in the United States. The selection-effect explanation Cofnas offers is real but partial. Selection effects amplify whatever potential exists, but they cannot create capacities that are not present. If the selected Nigerian sample exceeds the white American mean, then the selected sample’s underlying capacity exceeds the white American mean. The hereditarian frame should welcome this rather than redirect away from it, because it shows that the relevant variation runs along selected-versus-unselected lines rather than racial-group lines. Cofnas does not pursue the implication because the implication undermines the cleaner racial-group story his framework wants to tell.
The Harvard 0.7 percent figure deserves separate attention because Cofnas cites it across multiple interviews and treats it as decisive. The figure comes from Harvard’s internal modeling during the affirmative-action litigation and represents a counterfactual prediction about what the admitted class would look like if academic credentials alone determined admission. Cofnas treats this as evidence that hereditarian disparities at the very high tail are massive. The figure is real but the interpretation is loaded. Harvard admits roughly two thousand students per year out of an applicant pool already filtered through self-selection, college counseling, standardized testing access, and so on. The 0.7 percent counterfactual reflects the tail of a tail of a tail. It tells you almost nothing about population-level capacity distributions. It tells you something about the cumulative effect of multiple sequential filters that compound across the distribution. Cofnas’s framework treats it as a clean measurement of underlying capacity. It is not.
The career-and-future section at the end is the most candid moment in the interview, and tracks closely with the candor in the Razib interview. Cofnas wants to remain in mainstream academia. He hopes to reform institutions from within. He understands that this involves struggle. The candor is striking because the strategic vision it implies is in tension with the strategic vision he articulates elsewhere. The hereditarian-revolution essays call for purging departments, building parallel institutions, and morally discrediting the existing academy. The career statements call for staying in mainstream academia and reforming it from within. These cannot both be the strategy. Either institutional reform from within the existing structure is possible, in which case the moral-discrediting language is too strong, or the existing structure is too captured to reform from within, in which case the career strategy is built on a hope the framework should have already foreclosed. Cofnas does not resolve the tension. He holds both positions because his coalition needs both.
Bock brings up the law-professor anecdote about football admissions for a reason that becomes clear on reflection. He is testing whether Cofnas will endorse the position that admission should be by merit alone, including for athletics. Cofnas does not take the bait. He stays focused on the racial-disparity question. The interview is structured throughout this way. Bock floats positions Cofnas could endorse and watches whether he does. Cofnas declines most of the bait, which is correct litigation hygiene but also a tell about the coalition position he is maintaining. He does not want to be on record endorsing positions that his lawsuit-related counsel might advise against, and he does not want to be on record endorsing positions that would tie him to the populist hereditarian coalition he openly disavows in the Boyce interview. The careful surface costs the interview some of its substantive depth.
Cofnas is operating at the intersection of four coalitions whose memberships do not overlap. Mainstream academic philosophy provides his credentials and his employment context. Heterodox-academic free-speech advocacy provides his lawsuit funding and his media platform. The hereditarian intellectual coalition provides his substantive content and his readership. The libertarian-civil-liberties coalition provides his rhetorical framing and his theoretical foundation. Each coalition wants something different from him. Mainstream academic philosophy wants careful prose and acknowledgment of philosophical complexity. Heterodox academic advocacy wants principled commitment to free inquiry independent of viewpoint. The hereditarian coalition wants substantive empirical claims about race and intelligence. The libertarian-civil-liberties coalition wants opposition to both wokeism and to its rougher right-wing critics. The interviews work as well as they do because Cofnas is genuinely talented at maintaining all four registers simultaneously. The framework cracks at the points where the coalitions make incompatible demands, which is why metaethical anti-realism coexists with moral confidence in political claims, why career hopes for institutional reform coexist with rhetorical demands for institutional purge, why the Nietzschean self-identification coexists with the appeal to Christian moral concerns about disparities, and why the rejection of current race realists as unfit to govern coexists with the prediction that hereditarian victory will produce a better governing class.
The Cofnas framework is not a coherent philosophical position. It is a coalition equilibrium that maintains itself by holding contradictory commitments in suspension. Pinsof would predict exactly this structure. Trivers would predict that the position-holder would not recognize the structure as such. Becker would predict that the position-holder would experience his own work as cosmic-stakes truth-telling rather than as coalition-position-management. Turner would predict that the proceduralist defenses Cofnas offers cannot ground themselves in their own procedures.

‘The Taboo of “Race Realism” | with Dr. Nathan Cofnas’ (Mar. 4, 2024)

This shows Cofnas before the persecution narrative crystallized. The Cambridge controversy had just broken. He had not yet developed the polished talking points. The framework is being assembled in real time, and the assembly seams are visible in ways the later interviews paper over.

In March 2024 Cofnas’s coalition position is unstable in a different way than it became later. He still had the Leverhulme position, the lawsuit had not been filed, and the Substack project had only two posts. He depended on Cambridge institutional standing, on the small philosophy-of-biology coalition that took human variation seriously, and on whatever readership the Substack would build. He needed to attract sympathetic right-leaning intellectuals without losing the philosophy-of-biology peers who were watching to see whether he would handle the controversy with academic dignity. His coalition signals here are heavier on philosophical sophistication, lighter on populist edge, and noticeably uncertain about what the Substack project would become. He had less to lose than he later did because he had less established than he later did, and the prose reflects that.

In the metaethics opening, Cofnas spends fifteen minutes explaining that natural selection does not track moral truth, that our moral intuitions are products of evolutionary pressures unrelated to moral reality, that this debunks naive moral realism. He then spends the rest of the interview confidently asserting moral claims about wokeism, about the rightness of free speech, about what we ought to do with the equality thesis, about the proper organization of multiethnic societies. The contradiction is structural and goes unnoticed by the interviewer. If natural selection has given us moral intuitions that do not track any objective moral reality, then Cofnas’s confident moral claims about wokeism being a wrong response, about the moral imperative of telling the truth, about what societies ought to do, all sit on the same evolutionary substrate that the metaethical argument was supposed to debunk. He cannot have it both ways. Either moral claims have purchase, in which case the debunking argument was overstated, or they do not, in which case the rest of the interview is empty rhetoric.

This is the problem at the center of every position Cofnas has taken across many interviews. The hereditarian revolution requires that wokeism be wrong morally. Wokeism cannot be wrong morally if moral claims are evolutionary epiphenomena. The framework needs robust moral realism to do the work Cofnas wants and his own metaethics undermines that realism. Cofnas is not aware of the contradiction because the contradiction would dissolve the project. The mind protects the project from the recognition.

The path-of-victory section is the second most revealing passage. Boyce asks how the hereditarian wins. Cofnas’s answer is that hereditarians have truth on their side, that hereditarian intellectuals at universities and in prominent positions know the truth privately, and that what they need to do is show up to the fight and pay costs. He cites the historical example of communists and liberals who paid real costs and eventually won. The framing is interesting. Cofnas casts himself as the equivalent of the early-twentieth-century communist or liberal activist, willing to lose his job for truth, paving the way for eventual victory. The frame requires that the hereditarian position be true in the way that communist or liberal positions were true, and that institutional capture is achievable through the same mechanisms. Both assumptions are doing heavy lifting. Becker’s hero system frame fits perfectly. Cofnas has located himself in a cosmic narrative where his sacrifices contribute to eventual hereditarian institutional capture. The role rewards the role-holder regardless of whether the underlying empirical and political assumptions are correct.

The race-realist coalition exchange is sharper here than in later interviews. Cofnas openly says that current self-identified race realists are mostly haters and conspiracy theorists who would elect a worse president than the current one. He says he does not want to be ruled by the people currently identifying as race realists. This is a frank admission with serious consequences for his strategic vision. The hereditarian revolution he calls for would mean transferring institutional power to a coalition that, by his own assessment, currently consists mostly of people unfit to hold institutional power. He resolves the difficulty by predicting that if hereditarianism became accepted, a different and better class of people would adopt it. The prediction is unsupported. There is no historical example he cites of an unpopular position becoming popular and thereby attracting better adherents. The standard pattern is the opposite. Movements attract their adherents and then evolve to fit those adherents. The current race-realist coalition is what hereditarianism actually attracts. Cofnas is betting that institutional adoption would change the coalition’s composition. The bet has no evidentiary basis.

The Nietzsche section shows where Cofnas’s intellectual identity sits when he is talking with someone he respects but does not need to defer to. He identifies as Nietzschean, rejects Christianity, rejects the conservative defense of tradition, embraces transhumanism and genetic engineering. This is a more candid self-presentation than appears in any of the other five interviews. With Razib he gestured at his future book on the evolution of morality. With Boyce he says explicitly that morality is subjective and value is created by us. The Nietzschean self-identification is doing the same work the metaethical opening was doing. It positions Cofnas outside the moral framework that wokeism operates within, which means he can criticize wokeism without having to accept the moral premises that ground its critique of him. The strategy is structurally similar to what Strauss did. Want the benefits of moral confidence in attacking the existing order without paying the moral-realist costs that would expose your own position to symmetric attack.

The eugenics passage near the end is where Cofnas’s transhumanism shows clearest. He says embryo selection will be used disproportionately by intelligent and foresightful people, that this will produce more inequality, and that he is not concerned about the inequality. Boyce raises the standard worry about whether the smart will oppress the unintelligent. Cofnas’s answer is that the smart could already oppress the unintelligent now and do not, so why expect them to start when the gap widens. This is the framework’s optimistic register. Notice that the same framework operates in his analysis of wokeism in the pessimistic register. There the smart leftists are using their cognitive advantages to dominate institutions, suppress dissenters, and impose ideology on populations that do not share their values. Cognitive elites cannot both reliably dominate when they are leftist and reliably restrain themselves from domination when they are hereditarian. The asymmetry is doing coalition work, not analytic work.

The Christianity section is unusually weak. Boyce makes the substantive point that Christianity has demonstrated capacity to scale across intelligence levels and bind diverse populations into functioning societies, and asks whether Cofnas’s preferred framework can do similar work. Cofnas’s answer is that any successful ideological system needs a way to appeal across the intelligence spectrum, that Christianity worked because it offered both simple moral messages for ordinary people and complicated theology for intellectuals, that leftism has some of this and libertarianism does not, and that he prefers Nietzscheanism. The answer concedes Boyce’s point without seeming to notice it. If a successful ideological system needs to scale, and Nietzscheanism by Cofnas’s own admission cannot scale because Nietzsche recognized his ideas would not be widely understood, then the framework Cofnas prefers is by his own criteria not viable as a coalition-binding ideology. The hereditarian revolution he calls for cannot be Nietzschean because no scalable society could run on Nietzscheanism. It would have to be something else. Cofnas does not say what. The gap between his preferred personal philosophy and his proposed political project goes unaddressed.

Boyce is warm, intellectually curious, and inclined to take Cofnas seriously without pressing hard. The combination produces a relaxed Cofnas who reveals more than he intends to. The metaethics, the Nietzschean identification, the frank admission about current race realists being unfit to govern, the transhumanist enthusiasm, all surface here in ways that the later interviews suppress. Whether the suppression in later interviews is conscious or whether the framework itself adjusted under pressure, the comparison is informative. The Cofnas of March 2024 is more philosophically interesting and politically less coherent than the Cofnas of subsequent appearances.

Cofnas is a Nietzschean transhumanist who needs to operate within a Christian moral framework to make his political project legible. He is a metaethical anti-realist who needs robust moral realism to ground his critique of wokeism. He is a credentialed academic who needs the legitimacy that academic credentials provide while attacking the institutions that grant those credentials. He is a hereditarian who admits the current hereditarian coalition is unfit to govern but predicts a better hereditarian coalition will emerge if hereditarianism becomes mainstream. Each of these tensions is structural, not incidental. They are what holds the project together by holding contradictory commitments in suspension. Alliance Theory angle predicts exactly this structure. Coalition members hold beliefs that serve coalition function, not beliefs that cohere into a single consistent worldview. The function is to maintain Cofnas’s position at the intersection of multiple coalitions whose memberships do not overlap. The intersection is small but it is where his work lives.

The Boyce interview shows the assembly seams. The later interviews show the polished surface. Both are useful. The seams tell you what the polish is concealing.

‘Are Smart People Superior? A Reply to Noah Carl and Charles Murray’ (Apr. 26, 2024)

The strongest move is forcing Carl into a concession. Once Carl writes that “all else being equal, higher IQ equates to greater moral worth,” he has handed Cofnas the conclusion. The rest is bookkeeping. Cofnas catches the concession and presses it.
The weakest move is treating Murray’s position as incoherent rather than as a threshold view. Kantian dignity attaches to rational agency above some minimum and does not scale. Christian dignity attaches to image-bearing and does not scale. Both are defensible positions with long pedigrees. Cofnas waves them aside with “there is no indication Murray would want to take this position,” which is an assertion, not an argument. The threshold view deserves direct engagement. It does not get any.
His moral anti-realism does most of the heavy lifting. Once moral facts reduce to conventions expressing collective values, worth reduces to whatever a community values, and communities value intelligence. But the same logic produces parallel conclusions for beauty, charisma, athletic skill, family connection, and tribal membership. The argument generalizes. People vary in worth along every valued dimension. That conclusion may be correct. It also makes the focus on intelligence look like a coalition signal rather than a philosophical result. Why this trait, written up at this length, in this venue, by this author? The essay does not ask.
The drowning case smuggles instrumental considerations through the back door. We save children over adults because they have more life ahead. We save parents over the childless because of dependent suffering. The convention of women and children first comes from maritime tradition, reproductive logic, and heuristics about physical vulnerability. These are forward-looking calculations and inherited customs, not declarations of differential intrinsic worth.
The von Neumann move is sharp. Cofnas catches Murray in a rhetorical setup that exploits self-partiality rather than the structure of worth. Reframed as “von Neumann versus a random person,” the answer flips. Clean analysis.
The political equality section concedes more than Cofnas seems to notice. If the law looks past differences in worth, that practice carries a substantive moral commitment. It says: in the political sphere, treat people as if they had equal worth, even when they do not. That is close to the Murray position, relocated to the level of legal architecture. The egalitarian fiction has work to do in the world. Cofnas calls it a fiction. Murray might respond that the fiction counts for more in practice than the underlying truth, since the underlying truth is largely unactionable without producing the caste system Cofnas wants to avoid.
The closing exhortation to high-IQ children sounds noble. It also produces an asymmetry. What message goes to low-IQ children? Cofnas does not say. Christian egalitarianism has an answer: every soul has equal dignity before God, and your worth does not turn on your cognitive endowment. Cofnas leaves low-IQ people with diminished worth and no comparable consolation. The George Motz example aims at this gap, but Cofnas concedes Motz is probably “well above average” in IQ. The example does not show that low intelligence is compatible with high worth. It shows that high-but-not-stratospheric intelligence directed at a craft can produce worth. Different claim.
A coalition reading helps here. Cofnas writes from inside the hereditarian camp, correcting Murray and Carl and defending Kirkegaard and Kershnar. The status game is rigor and willingness to follow the argument where others flinch. Murray’s softer line is the coalition’s public-facing posture. It holds ground in mainstream debate. Cofnas occupies the internal vanguard, and his complaint is that the public-facing posture concedes too much. Read this way, the essay is partly a discipline document for the camp. Stop apologizing. Stop pretending the conclusion does not follow.
The piece gains force from its honesty about premises. Cofnas tells you he rejects moral realism. He tells you he rejects Christian egalitarianism. He tells you what he thinks the truth is and does not hide behind hedges. That is a virtue. The cost is that readers who do not share the premises have no reason to share the conclusions, and the essay does little to bring them across.

‘Race, Wokism, and Academia with Amy Wax’ (Jun. 7, 2024)

Cofnas says intelligence has intrinsic value. Wax pulls him back. Excellence is one thing, moral worth another. Ordinary people can live virtuous lives. The two can come apart. Wax cites Charles Murray and Noah Carl as authorities for the separation. She knows what Cofnas’s position costs the coalition. If hereditarianism becomes the claim that smart people are morally better, the religious Right walks away, the working-class Right walks away, and the movement collapses into a thin band of coastal IQ enthusiasts. Wax cannot afford that. Her case at Penn rests on a coalition that includes the religious Right.
Apply the four questions to Wax. Status, income, protection: tenure at Penn law with the September 2024 sanctions hanging over it, the lawsuit dismissed by the federal judge in August 2025, Federalist Society networks, the conservative legal scene, podcast circuit, Manhattan Institute orbit, donor support that pays the legal bills. Allies to attract or retain: Murray, Sowell, Hanania, Coleman Hughes, the religious Right that needs virtue language, mainstream conservative intellectuals who tolerate race realism when it stays well behaved. Membership signals: hard realism on group differences, K-12 as the front, anti-feminization claims about Academia, marriage and children as the conservative answer, the careful separation of excellence from moral worth, defense of Western Civilization as a unitary inheritance. What she gives up if she changes: the lawsuit narrative, the religious Right’s permission slip for her hard realism, the seat at the conservative legal table.
Apply the same questions to Heterodox Academy as both speakers describe it. Status, income, protection: foundation funding from center-left donors, university affiliations, the Haidt brand, the speaker circuit. Allies to attract: moderate liberal academics, soft critics of cancel culture, donors who want viewpoint diversity without trouble. Membership signals: the careful distance from race realism, the calm reasonable voice, the position that we are the adults in the room. What they give up by defending Wax or Cofnas: access to mainstream institutions, donor confidence, the position between the camps. Heterodox Academy chose. Wax and Cofnas note the choice. The choice was coalition rational from the inside even if it looks like betrayal from where Wax and Cofnas sit.
The Hanania exchange is coalition strategy expressed as empirical dispute. Hanania’s “shut up about race and IQ” piece tells Cofnas the topic is bad coalition tactics. Cofnas counters that arguments work over time, citing biblical literalism’s retreat. Both men hold race-realist positions. The dispute is about timing and audience, not substance. Cofnas treats Hanania as having capitulated. Hanania treats Cofnas as overplaying a hand the coalition can win without showing. Both might be right at different scales.
Wax’s contribution on Hanania is more careful. She grants him the proximate cause of woke institutional behavior in civil rights law as enforced. She grants Cofnas the deeper cause in the equality thesis. The split lets her hold the coalition together by giving each man partial credit. This is what senior figures do when junior figures squabble in public. The framing also positions Wax as the synthesist, the figure above the dispute, which is a coalition gain in itself.
Trivers shows up across the conversation and inside neither speaker’s self-understanding. Wax says her mindset has been influenced by knowledge of empirical reality. Cofnas says smart people are disproportionately woke because they followed the argument, and that he and his coalition followed the argument one step further. Both treat themselves as standing outside the self-deception field that captures everyone else. The position is structurally identical to the position of the woke academics they criticize, who also believe they followed the argument and the others have not. Two coalitions, one epistemic posture.
Turner pushes against the essentialist core. Wax slides between hard realism and soft realism as the conversation requires. She wants the cutting edge of group differences in cognitive ability and the hedge of cultural and social explanations. The two positions do different coalition work. Hard realism recruits the race-realist scene. Soft realism preserves the religious Right’s dignity claims about virtue and the average person. Wax holds both because each is needed for a different audience. The Western Civilization she wants K-12 to teach is an essentialist category. The feminization thesis about Academia treats women’s left lean as a fixed feature of female participation rather than a contingent product of specific historical conditions she notes she does not understand.
The stupid smart people line cuts against Cofnas’s framework and Wax does not press the cut. She concedes that many high-IQ people hold what she calls silly Lefty ideas. If high IQ does not track holding correct positions, then the argument that smart people followed the wokeness argument because they are smart loses force. Cofnas does not pick up the thread. The concession is in his interest to ignore. The coalition keeps the claim that wokeness captured the smart and the claim that the smart are often wrong, because each does work in different contexts.
The K-12 pivot is where Wax becomes the political operative and Cofnas remains the intellectual. Wax tells young men to get married, have children, and run for school board. Cofnas does not engage with the practical program. He returns to the claim that the right has too little human capital to take the universities. Wax’s program is coalition building. Cofnas’s program is coalition theorizing. Both have a place. Wax knows hers and gently tells Cofnas that his part of the work is upstream of the part that wins elections and changes K-12 curricula.
The reproof on intrinsic value is the same move at the meta level. Cofnas wants the hereditarian revolution to mean intelligence carries moral weight. Wax wants the hereditarian revolution to remain compatible with a virtue ethics that licenses ordinary lives. Cofnas’s version flatters his audience and shrinks it. Wax’s version dilutes the message and grows it. Wax has been at this longer. Her version is the one that survives contact with K-12 reform and a Federalist Society fundraiser.

‘Natural selection requires no teleology in addition to heritable variation in fitness’ (Aug. 7, 2024)

Cofnas does useful conceptual housekeeping. Lewontin’s three conditions (variation, differential fitness, heritability) do technically cover artificial selection, intelligent design, and forward-looking orthogenesis. If you want a formulation that captures what was new about Darwin, you need something more. The “no teleology” addition fixes a real bug.
The historical argument carries the paper. Darwin built his case on the analogy with breeding while insisting that what happens in nature lacks the guiding hand. He treated Lamarckian inheritance as an alternative to natural selection, not a special case of it. Cofnas reads this correctly. Huxley’s response to Kölliker captures the point: the Origin killed teleology in the ordinary sense.
The agential and natural teleology split clarifies things. Agential teleology requires a mentally represented endpoint. Natural teleology means a forward-looking force without a mind behind it. Cofnas formalizes both with probability conditions. The natural teleological mutation definition (mutation likelier when adaptive, not because of mutagenicity or inherent bias) is clean. So is the natural teleological selection definition, where an intermediate trait gets selected because of where it leads.
The paper has trouble in three places.
First, the Boehm move. Cofnas classifies Boehm’s “social selection” theory as teleological unconscious selection because the egalitarian coalitions had a blueprint of desired social arrangements. But his own treatment of hunters and deer says that human interventions producing evolutionary side effects do not count as teleological if the human aim is not to drive evolution. The egalitarian coalitions wanted compliance with norms in their own generation. They did not aim at the evolution of conscience. By the deer-hunter logic, Boehm’s case should count as standard natural selection. Cofnas might reply that the trait targeted (compliance with norms) was selected on the basis of a mental representation. But that just means the selectors wanted compliance, not that they wanted heritable compliance. The line between his two cases looks unstable.
Second, the “not entirely the result of” qualifier in his probability conditions creates a measurement problem. If a process is partly agential and partly natural, when does it stop counting as natural selection? A breeder who sometimes lets nature take its course produces some adaptations through agency and some through unguided forces. Cofnas needs an answer to mixed cases that goes beyond “not entirely.”
Third, sexual selection. He calls it “normally” nonteleological because animals lack an explicit mental representation of an ideal mate. But preference is a forward-looking psychological process. The peahen does not need a mental picture of a long tail to be preferring long tails. If preference counts as agential telos, much of sexual selection becomes teleological. If preference does not count, the line between conscious and unconscious selection in his earlier discussion gets harder to draw.
Cofnas grants that human cultural evolution is shot through with teleology, and that designed institutions can create selection pressures that change humans biologically. This is the niche construction point made stronger. The standard formulation lets you call all of this natural selection. His formulation forces you to pull the strands apart: which adaptations come from blind forces, which from designed pressures, which from a mix. For thinking about how elite institutions shape the people in them, this is the more useful framing.
The practical payoff is small for working biologists, who already treat artificial selection as a separate force in textbooks. The payoff is larger for philosophers who want their formulations to do conceptual work, and for anyone analyzing cultural and institutional change where designed pressures and blind forces both operate. A worthy small contribution. The bug it fixes is real. The fix needs another round of work to handle mixed cases.

‘Thomas Sowell’s Wishful Thinking about Race’ (Aug. 12, 2024)

Cofnas catches Sowell in clear errors. The Jensen misquote about Appalachian inbreeding shows Sowell read carelessly or filtered through his preferred reading. The “Ulster County” slip suggests he never checked the geography of the very ancestors he treats as load-bearing for his theory. The Cicero quote is doctored. The Dunbar IQ figures lack provenance. These are scholarly failures, and Cofnas is right to point them out.
The logical points also land. Sowell repeatedly slides between “not all Whites are the same” and “therefore hereditarianism is wrong,” when no serious hereditarian ever claimed all Whites were the same. The selective migration explanation for higher Northern Black IQ scores fits the pattern Cofnas describes better than the cultural absorption story. The Philadelphia mulatto data cuts against Sowell more than for him, since the mulatto-Black gap within the same Northern city tracks something other than exposure to White Yankee culture.
But Cofnas has his own weaknesses.
The framing of three options, racism or genes or culture, is too clean. Geography, institutions, capital accumulation, family structure, religion, technology, geopolitics, and timing of contact with industrial economies all shape group outcomes. Both Sowell and Cofnas treat culture as a residual category, and that flattens the analysis on both sides.
His Jewish counterexample undermines his own position more than he sees. Brigham in 1923 measured Jews as low-IQ. By 1970 Jews dominated American intellectual life. Cofnas explains this away by saying the early tests were flawed, the subjects were immigrants, the language and culture were strange. Fine. But that concession damages the strongest version of his hereditarian claim. If testing artifacts and environmental conditions can move a group’s measured IQ that far in fifty years, the current Black-White gap might also reflect testable environmental and cultural distortions. He cannot have it both ways.
The political argument at the end is the weakest part. He claims hereditarianism is the only way to defeat wokeism. People can accept innate group differences and still demand redistribution, equal protection, and institutional reform. The moral premises shift independently of the empirical premises. Murray acknowledges this. The hereditarian who thinks the genetic case forces a political conclusion has already made a leap Cofnas does not justify.
The coalition layer operates strongly here. Cofnas writes as a member of a particular tribe, the heterodox right that wants to displace mainstream conservatism. Sowell sits at the heart of that mainstream, Hoover, National Review, Heritage. Cofnas attacks Sowell partly to signal that the older establishment failed to protect the right ideologically, and a younger, more empirical, more willing-to-cross-lines movement should take the lead. Notice how he closes. We have two choices, win with hereditarianism or lose. That is recruitment language. Join my faction or surrender.
Sowell’s culturalism has its own coalition logic. It lets respectable conservatives criticize Black behavioral patterns without touching the genetic third rail. It permits discussion of crime, family breakdown, and school failure, while keeping Black people morally agentic rather than biologically fated. Whether or not it is true, it serves a movement that wants to discuss racial disparities without inviting accusations of racism. Cofnas wants conservatives to give up that rhetorical advantage and accept the costs of saying the harder thing.
The honest scholarly position is that the question is hard and the evidence underdetermines the answer. Group differences in measured cognitive performance are stable. Selective migration explains some of it. Environmental and testing artifacts explain some of it. Genes might explain some of it. The Flynn effect shows that large environmental shifts move scores massively across generations within the same population. Adoption studies show within-family environmental effects shrink with age. Brain size differences exist but their cognitive significance is contested. The evidence does not point cleanly to any single answer.
Cofnas writes as if it does because his coalition demands certainty. Sowell writes as if culture explains everything because his coalition demands a non-genetic story. Both men are advocates more than investigators on this question.
Sowell’s Sixty books is a warning sign. No one writes that many books and maintains rigor. His admirers stopped checking his footnotes a long time ago, and he stopped expecting them to. The same risk applies to Cofnas, who writes with the confidence of a young scholar who has not yet had his major claims dismantled in print.

‘Wokism Is Just Beginning’ (Oct. 9, 2024)

Cofnas argues from a single premise: wokism follows logically from the equality thesis. Because elites believe races have identical innate distributions of potential, persistent disparities can only be environmental, so disparities create a moral emergency that demands ever more aggressive intervention. From that premise everything else follows. The only escape is hereditarianism.
His point about cancellation rates declining partly because the dissidents have already been purged catches something. The institutional filter has done its work. Harvard renaming diversity statements to service statements shows how surface change covers continuity. The First Crusade analogy works as a corrective to triumphalist past-peak-woke essays. Movements often look spent at the moment they consolidate.
The analytic frame breaks down at the load-bearing point. Cofnas treats wokism as deduction from premise. Coalitions do not work that way. Moral vocabularies are not derived from empirical claims. They are coalition technologies. The empirical claim follows the coalition, not the reverse. If hereditarianism became scientifically settled tomorrow, the coalition that benefits from antiracist vocabulary might not dissolve. It might find new vocabulary. The same status hierarchies and the same rents might persist under a different rationale.
This is where Pinsof cuts hardest against Cofnas. Cofnas writes as if persuading elites of group differences might deflate the whole structure. But elites have stronger reasons to hold the equality thesis than to follow evidence. The thesis legitimates their authority over redistributive institutions, their gatekeeping power in hiring, and their moral standing in their professional networks. Asking them to abandon it asks them to give up status, income, and belonging at once. My four questions apply directly to anyone Cofnas hopes to convert: who does this elite rely on for status, income, and protection, who must he attract or retain as an ally, what signals mark his coalition, and what does he give up if he changes position? The answer for almost every elite is loss on all four.
There is also a Turner problem. Cofnas wants to defeat one essentialism by installing another. The blank-slate position treats race as social construction with no biological reality. The hereditarian position treats race as biological essence that explains group outcomes. Both errors come from the same family. Turner’s critique of Vermeule and Deneen applies here. You cannot reverse-engineer a working public order from a thin metaphysical claim. Replacing the equality thesis with a hereditarian thesis does not produce the political order Cofnas wants. It produces a different coalition with its own grievances, its own enforcement apparatus, its own essentialism.
Becker fits both sides. Wokism is a hero system. So is hereditarianism as Cofnas presents it. Both organize meaning around opposition to a great evil. Both promise that following the truth redeems the world. The structural symmetry should give pause to anyone tempted by Cofnas’s confidence.
Cofnas presents himself as the one who sees clearly while the equality theorists deceive themselves. Trivers predicts that members of any coalition believe they alone have escaped the bias their opponents suffer. The certainty is the tell.
The empirical predictions have problems too. Generational turnover does not produce steady ideological intensification. Movements exhaust themselves through overreach, internal contradiction, and material costs that hit even the believers. The period since the essay appeared shows a more mixed picture than Cofnas predicted. Substantive corporate retreats from DEI. Court rulings against racial preferences. Federal action under the second Trump administration that goes well beyond what Cofnas treats as the ceiling for any conservative win. None of this proves Cofnas wrong about the long arc. It does show that his linear projection model is too simple.
The strongest version of his argument is institutional. Boomer and gen X gatekeepers hold many institutions, and their replacement might shift the median position leftward. The cohort data on cancel culture support stands out. But that is a forces argument, not the logical-deduction argument the essay frames as primary.
What Cofnas gets right: complacency about wokism’s decline is unwarranted. What he gets wrong: he treats an ideology as a syllogism, and proposes a solution his own framework predicts cannot work.

Stanford University Classical Liberalism Seminar – Nathan Cofnas – (Sep. 5, 2024)

The Stanford talk shows the equality thesis under stress test from a friendly room. The room is friendly. Classical liberals who run a seminar series, audience members already sympathetic to free speech. They still push back. The pushback is informative because it exposes the load-bearing claims Cofnas needs and the joints where the weight is not landing.
Apply the four questions to the audience. The Classical Liberalism Initiative needs to remain a credible institutional home for free speech work at Stanford. Status, income, protection: foundation funding, Stanford affiliation, donor expectations, the credibility of running a seminar that hosts speakers like Cofnas without becoming a vehicle for hereditarianism. Allies to attract: classical liberals across the political spectrum, donors who care about academic freedom but not about race science, faculty who tolerate the seminar series. Membership signals: rigorous engagement, careful pushback, refusal to swallow strong claims without evidence, the position that classical liberalism remains a viable third option between woke left and reaganite right. What they give up if they accept Cofnas’s framing: the third option position. If wokeness follows from the equality thesis plus Christian morality, classical liberalism collapses into either capitulation to wokeness or hereditarianism, and the seminar becomes a recruiting station for one side.
The room knows this. Watch the resistance. The host interrupts to dispute that classical liberals are not a movement. He names the Reaganite Conservative scene, the National Review crowd, thousands of people. Cofnas responds that the question is why those movements fail to attract elites in serious numbers. The exchange is coalition negotiation. The host wants the seminar to remain a credible classical liberal venue. Cofnas wants to recruit the classical liberals to the hereditarian position.
The pushback on intellectual coherence comes from multiple directions. One audience member asks whether Cofnas would apply the same logic to communism, which attracted intellectuals more than common people. Cofnas redirects, treating communism as a different scenario rooted in ancient egalitarian impulses. The redirect protects the framework. If wokeness draws elites because it is intellectually coherent, communism drawing elites should require the same explanation. Cofnas declines the parallel because it would force the framework to predict that twentieth century elites correctly followed an argument to mass starvation.
A second audience member pushes harder. Just because intellectuals accept an idea does not show the idea is intellectually coherent. The point cuts to the joint. Cofnas’s argument needs the inference from elite acceptance to intellectual coherence. The audience member offers alternative drivers: religious nature seeking utopian substitutes, social pressure, professional incentives, the Henry James-style resentment of intellectuals who watch billionaires succeed without understanding anything. Cofnas grants that other forces operate. He maintains that the equality thesis is the distinctive driver of what he calls wokeism. The retreat is rhetorical. If many forces produce the phenomenon, the strong claim that wokeness follows logically from the equality thesis loses force.
A third pushback comes from the host on the move from equality thesis to wokeness as a logical consequence. The host disaggregates the equality thesis. Does it mean equal genetic distribution at birth? Equal cultural production by age eighteen? Equal preferences and choices that produce few women on oil rigs? Cofnas does not separate the components. The argument needs the strong version because the strong version is what generates the moral emergency. The weak versions, equal moral worth or equal legal treatment, do not produce the panic that drives the wokeness Cofnas wants to explain.
The Tucker Carlson section shows Cofnas’s stereotype accuracy claim doing real coalition work and at the same time burning his bridge to the populist Right. Carlson sniffing bread in a Russian supermarket and interviewing a Hitler apologist becomes evidence that the right is intellectually inferior. The audience laughs. The use of Carlson as a representative figure of the right is exactly the move Cofnas would reject if a left coalition used Robin DiAngelo to represent the left. The asymmetry shows Cofnas selecting the worst of the populist right and the smartest of the woke left and treating the comparison as evidence about the populations rather than about the selection.
Cofnas treats himself as the man who followed the argument one step further than the woke elites who followed it most of the way. The audience members treat themselves as smarter than both Cofnas and the woke elites because they see through the move from elite acceptance to intellectual coherence. Each position has the same epistemic posture. Each treats itself as standing outside the self-deception field that captures everyone else.
Turner pushes against the essentialism of Christian morality as Cofnas uses it. Cofnas treats Christian morality as a unified inheritance that produces wokeness when fed the equality thesis. The audience member who asks about non-Christian countries with under-represented minorities flags the problem. Indians worry about Dalit under-representation. African countries worry about ethnic representation in elite positions. If the same pattern shows up across religious traditions, the Christian morality genealogy explains less than Cofnas claims. He grants that the problem of multi-racial societies generates the taboo across cultures. The grant shrinks the role of Christian morality in the framework, which is the load-bearing premise that makes wokeness specifically Western.
The 0.67% Harvard number does specific coalition work. Cofnas needs the audience to feel that meritocratic admissions would produce a Harvard with almost no Black students. The number does the work. The framing assumes that current Black admission to Harvard reflects affirmative action rather than the joint operation of test prep, legacy, geographic distribution, athletic recruitment, and the actual right tail of a population of forty-two million. Harvard’s internal modeling produced the 0.67% under specific assumptions about academic-only criteria. Whether those assumptions match what classical liberal admissions would actually optimize for is a separate question. The audience does not press the assumptions because pressing would extend the argument into territory most of them prefer not to enter.
The radical action conclusion is where Cofnas’s framework reaches its operational form. He suggests a quota system might be necessary in the short to medium run to clear out certain groups from elite positions if the hereditarian revolution succeeds. The room goes quiet on this. The classical liberals in the audience cannot endorse a quota system run in the opposite direction from the current one. Their classical liberalism precludes it. Cofnas offers it because his framework requires some account of what comes next, and the implications of the framework are not soluble in classical liberal commitments. The joint between Cofnas’s project and the seminar’s project shows here.
The atheist scientist who speaks at the end gives the strongest pushback. Smart people are not woke. Fake intellectuals are woke. Real scientists are anti-woke classical liberals. The claim is mirror-image to Cofnas’s claim. Cofnas treats elite acceptance of wokeness as evidence of intellectual coherence. The scientist treats his anti-woke peers as the real intellectuals and the woke as the fake ones. Both moves redefine intellectual to include people who agree with the speaker. The scientist also names the gender point, which does not fit Cofnas’s framework cleanly. If wokeness is significantly gendered, with university-age women much more woke than men, then the equality thesis plus Christian morality story has to share explanatory work with whatever produces the gender pattern. Cofnas treats feminization of institutions as a separate phenomenon. The treatment protects the framework at the cost of what the framework can explain.
The poll numbers Cofnas presents at the end are real. Two thirds of Zoomers in the US and UK support firing James Damore. Half of young academics support cancellation campaigns. A majority of Zoomers think living in fear of cancellation is a justified price to protect historically disadvantaged groups. These numbers describe a coalition that has captured the rising generation in elite institutions. Whether the coalition holds depends on conditions Cofnas does not fully model. World Values Survey data shows generational cohorts crystallize their values and rarely shift them later. If the Zoomers carry their views forward, the institutions stay woke. If conditions change, the views might shift. Cofnas needs the second possibility for his radical action plan to make sense and the first possibility for his urgency to make sense. He uses both at different points without resolving the tension.
The Cambridge personal note shows the cost the framework imposes on its primary advocate. Cofnas describes the Daily Mail fabricated quote, the Telegraph piece calling for his removal, the radio interview he calls the nastiest some people had ever seen. The British anti-woke press behaves the way the woke American press behaves. Coalition discipline operates on both sides. Cofnas is a target for the dominant coalition because of his hereditarianism and a target for the populist anti-woke coalition because his hereditarianism breaks the etiquette they need to maintain access to mainstream institutions. He has no coalition home that will fund him and protect him. The lawsuit is what he has instead of a coalition home.

‘The Hereditarian Revolution with Nathan Cofnas’ (Nov. 18, 2024)

Cofnas lays out his hereditarian revolution as historical inevitability. The Soviet Union collapsed suddenly. The Boomer gatekeepers will retire. Vance and Musk follow HBD accounts on Twitter. The young right already knows. Within several years the trajectory might change. This is a Becker hero system claim with the costume of historical prediction. Meaning attaches to participation in the inevitable shift. The denial of death runs through the framing. I might be at the head of the spear when the institutions get retaken.
Apply the four questions to the philosophy profession as Cofnas describes it. Status, income, protection: tenure, prestige hierarchies, peer review networks, university endowments. Allies to attract: woke Twitter as Cofnas calls it, the HR apparatus, donors who care about diversity statements, the credentialed graduate student pipeline. Membership signals: Rawls commentary, technical analytic puzzles, diversity statements written with sufficient sincerity, the homework-doer disposition that produces a 3.9 GPA without picking fights. What members give up if they change: the moral consensus that licenses the institution to claim authority over questions about the good life.
The BAP fight is coalition material rather than philosophy. BAP and Cofnas compete for an overlapping audience of young right-coded intellectuals. BAP offers a hero system rooted in body, glory, and warrior aesthetics. Cofnas offers a hero system rooted in martyrdom for empirical truth. The dork insult is coalition discipline aimed at recruits who might join either side. Cofnas counters that BAP has a Yale Classics PhD and has never fought in a battle. The counter is fair. It also shows how much both men rely on coalition signaling about what counts as masculine seriousness, the substance of which neither has adjudicated.
Trivers shows up across the interview without entering the speakers’ self-understanding. Cofnas says many prominent figures hold one view in private and another in public. He acknowledges he might lie about a low IQ score. He frames his perfect verbal score and his eighty-second percentile quantitative score as confirmation of his theory rather than as a result that might prompt examination of the test as an instrument. His pattern of incentives and his reading of his own scores run on parallel tracks he does not cross.
Turner pushes against the essentialist core of the argument. Cofnas hedges in places. Appalachian Whites differ from Massachusetts Whites. Selected African immigrants in the UK differ from population-representative African samples. The hedges cut against the strong hereditarian frame his audience wants and that the Daily Mail attached to his name. The Locke-to-Christianity-to-WEIRD genealogy of wokeness is itself an essentialist claim about Western moral psychology, transmitted through a single line of descent from 1690 through Henrich. Whether the genealogy survives the kind of hedging Cofnas applies to race categories is a question he does not raise.
The equality thesis as Cofnas defines it does theoretical work. He frames it as the claim that all groups have the same distribution of innate ability. Wokeness then follows logically from that premise plus Christian morality. The move treats the equality thesis as monolithic. The actual cluster of claims in academic and political discourse contains many distinct propositions: legal equality of treatment, equal moral worth, equal opportunity, equal distribution of cognitive ability across populations. Cofnas conflates the strongest version with the version most institutions hold. The slippage lets him present every defender of equal moral worth as a covert defender of equal distribution. The argument gains rhetorical power and loses precision in the same step.

‘Talking about Race Differences with Nicholas Wade’ (Dec. 20, 2024)

Wade’s most revealing line is the admission that he “punted” on IQ. He says it openly. A Troublesome Inheritance is hereditarian-curious about social behavior because the IQ debate was too radioactive even for a man who had Penguin Press behind him and three decades at the New York Times. Apply my four questions to Wade and the position selection becomes obvious. He depends on book sales, residual prestige, and the kind of mainstream-adjacent platform that Penguin still represents. He must attract liberal readers willing to consider hereditarianism in soft form plus right-leaning intellectuals tired of blank slatism. His coalition signals are the punt, the focus on social behavior, and the Penguin imprint that signals respectability. He might lose the residual social capital and access if he went all the way on IQ.
Cofnas pushes him on this and Wade holds the line. The exchange is polite because both men need each other. Cofnas needs Wade’s mainstream credibility to validate his own niche. Wade needs Cofnas to keep the conversation going since his book is twelve years old now and the energy in this space has moved to Substack figures.
The Diamond comparison is the sharpest moment in the conversation. Diamond holds the same theoretical position as Wade. Selection pressures in different environments produce different cognitive distributions. Diamond says Papua New Guineans are smarter than Westerners because their selection pressures were harsher. He says it in print, with no qualification, and wins every prize a book can win. Wade says something more careful, more hedged, and gets pilloried. The asymmetry tells you that the substantive content of the claim does not determine the response. The direction of the claim does. Diamond’s claim flatters the right people. Wade’s might not. Coalition rules apply.
The Henrich critique Cofnas raises is the most analytically productive moment. Henrich’s WEIRD framework says the church broke up European kin networks through marriage rules, and this produced the individualism that paved the way for the industrial revolution. The hole is East Asia. East Asia received industrial modernity quickly without going through the church program. If the church marriage rules do all the work, why does the receptiveness travel so cleanly across the Eurasian landmass to populations that never had cousin marriage bans? Wade and Cofnas both note that Henrich never mentions genes, and that the gap shows where the cultural-only story breaks down. This is one of those cases where hereditarian explanations might fill in what cultural ones leave open.
The Locke move is sharper and Wade does not pick it up. Cofnas wants to push the origin of race denial back from Boas to Locke in 1690, arguing the impulse for blank slatism predates Jewish intellectual influence by centuries. This undermines MacDonald more cleanly than the default hypothesis does. If Locke is the founder of liberalism and the founder of blank slatism and the founder of race denial in one package, then race denial is internal to liberalism rather than external Jewish corruption of liberalism. Wade pivots to a more ecumenical point about liberal political instincts and equality. The reason might be that Wade does not want a fight with the MacDonald audience that overlaps with his own readership in odd ways.
The hierarchy exchange is the philosophical core. Cofnas presses Wade. If trait distributions vary and some traits make individuals more capable, does that mean some individuals carry more worth than others? Wade resists. He shifts to group-level success. Some societies function better than others, so let’s talk about that. Cofnas keeps pushing and Wade concedes a little, then redirects again. Watch the move. Wade is a hereditarian who refuses to follow the implications past a particular line. Cofnas is willing to follow them, says intelligence might bear on individual worth, qualifies that this rarely shows up in practice, but holds the philosophical position. The asymmetry is generational. Wade is older and remembers when admitting any of this got you shut out from everything. Cofnas is younger and writes for an audience that already accepts the basic structure. He has less to lose by following the argument where it goes.
The covid section sits awkwardly in the rest of the conversation but earns its place. Wade’s article was rejected by every mainstream outlet he approached. He published on Medium because he had nowhere else to put it. The bulletin of atomic scientists reprinted it. A million page views followed. The mainstream press he had served for three decades had no place for his most consequential piece of journalism. The institutional gatekeeping failed, and the gatekeepers paid no price. They continue to refuse to publish the strongest evidence for the lab leak position even after the position became respectable. Becker’s hero system frame applies. Mainstream science journalism has built its identity around fighting misinformation, and the lab leak hypothesis got coded as misinformation. Once a claim sits inside the misinformation category, the evidence cannot get it out. The category does the work, not the evidence. Trivers on self-deception covers the rest. The journalists who refuse to cover the evidence are not lying to their readers in any way they could admit to themselves. They have convinced themselves the evidence is not there to cover.
Cofnas can host this conversation on YouTube in 2024 without it getting taken down. His earlier podcast got pulled. He kept making material, found a way back, and now sits in a niche that did not exist five years ago. The platform calculation has shifted. YouTube allows hereditarian conversations as long as the participants frame them with academic seriousness and avoid the more inflammatory adjacent positions. Both men perform this calibration throughout. Wade hedges on IQ. Cofnas attacks anti-Semites alongside woke censors. The coalition logic of the platform selects for this kind of careful presentation, and the men know how to deliver it.
Cofnas’s default hypothesis works for him in attracting the man who wants hereditarianism without anti-Semitism. That guy is a real coalition and Cofnas serves it well.

‘Was I Wrong about Woke?’ (Jan. 29, 2025)

Cofnas identifies a logic of wokism and shows how the logic produces the features. Most “what is woke” pieces fail at the door because they catalogue rather than explain. He clears that bar.
The model has a real weakness, though, and it sits at the foundation. Cofnas treats wokism as a set of propositions reasoned from premises. Coalition analysis treats ideology as signal. Beliefs mark allegiance. Premises get retrofitted to justify signals already in use. If the second picture is closer to right, the policy implication of his essay collapses. You cannot dewokify elites by refuting the equality thesis, because the elites did not arrive at the position by reasoning from it.
The timing problem points the same way. The equality thesis went mainstream with Boas a century ago. Wokism did not arrive in 1935 or 1965 or 1985. It came around 2012, intensified after 2014, and again after 2020. Cofnas says people finally noticed that civil rights legislation had not closed gaps. The claim does not hold. Steele wrote about Black underperformance in the eighties. The Bell Curve came out in 1994. The gaps were visible. Wokism arrived when smartphones, institutional feminization, Obama-era racial reframing, and credential-class demographic shifts aligned. Cofnas concedes these as facilitating conditions. His model treats them as secondary. They carry the load his premises cannot.
The Christianity claim is loose, and he half-admits it. He says wokism does not require Christianity per se, just any morality of equal treatment. So why call it Christian? Because the framing gives the ideology a two-thousand-year pedigree and lets him treat secular Westerners as Christians in denial. The egalitarian premise he has in mind comes from Enlightenment universalism, not Pauline soteriology. The Christianity that produced abolition and the Christianity that produces DEI training share vocabulary, not logic.
The Rufo critique has bite. Rufo’s counter-elite plan runs into the human-capital problem. There are not tens of thousands of right-wing scholars ready to staff a university takeover. Conservative culture has self-selected for grift and crankery long enough that the talent pool is thin. Cofnas is right about that.
His own solution has the same problem in reverse. The people willing to publicly champion hereditarianism are largely unsuitable as elite signal-bearers. Murray hedges because the cost of not hedging is exclusion. Cofnas lost his Cambridge position. His information campaign requires a coalition of credentialed scientists who accept the reputational hit. That coalition does not exist and cannot be summoned by demonstrating premises.
The colorblind regime point cuts the wrong way for him. He says colorblindness without hereditarianism sets up a second Great Awokening. Fine. But colorblindness with hereditarianism requires elite figures to publicly defend outcomes that look like apartheid by demographics. His own chart shows 0.55 percent Black representation at the 140 IQ band. Defending that publicly is harder, not easier, than defending it from the equality premise. The political weight of an avowedly hereditarian elite consensus is enormous. He has not accounted for it.
His tech-bro evidence is thin. Musk reposting HBD accounts is not evidence the tech right was won by hereditarianism. The tech right was already right-coded for reasons unrelated to race science: COVID lockdowns, regulatory friction, Twitter censorship, immigration policy. HBD discourse circulates in those circles because the coalition has formed, not because the coalition formed around HBD. Cofnas reads coalition signals as ideological conversion.
The strongest part of the essay is the Laurie Penny challenge. Most conservatives cannot define woke. The right’s inability to answer is the symptom of a movement that operates as reaction. Cofnas at least offers a definition that explains something. His definition is probably not fully right. But his critics have not offered better, and that is a real indictment of the anti-woke project he is criticizing.

‘The Controversial Science of IQ & Culture – Nathan Cofnas’ (Jan. 30, 2025)

The interview presents a contest about scientific truth. The contest is a coalition fight.
Cofnas claims hereditarianism describes reality. The University of Cambridge claims diversity, equity, and inclusion describe its values. Both claims do work in the world. They recruit allies, mark membership, license the holder to speak with authority. The truth question and the coalition question run on the same track.
Apply the four questions to Emmanuel College. Status, income, protection: Cambridge’s reputation, Russell Group standing, applicant streams from progressive sixth forms, philanthropic gifts contingent on DEI alignment. Allies to attract: students, junior fellows, donors who care about diversity statements. Membership signals: published commitments to equity values, response to incidents like Noal Carl, the public posture of inclusion. What they give up by retaining Cofnas: alignment with the broader sector, donor confidence, the shared vocabulary that lets them recognize each other as members of the same club.
Each side performs its hero system. Cofnas plays the heretic, a man who tells the truth at cost. The Becker move sits on the surface. He compares himself to Neven Sesardić facing communist Yugoslavia and to Galileo-style truth-tellers. The denial of death runs through the framing: I might be remembered for telling the truth that others would not. Emmanuel College plays a different hero system, guardian of inclusion, defender of the Black student whose presence Ferrari invokes. Both systems promise meaning through participation in something larger than the participant.
Cofnas says many prominent figures hold one view in private and another in public. He acknowledges he might lie about a low score. He notes his own incentive to share the verbal score and not the quantitative score. Yet he treats his own position as outside the self-deception field. He is the man who sees clearly. Gold accepts this framing.
The Daily Mail episode shows what proceduralism looks like in practice. The paper attaches quotation marks to a sentence Cofnas did not utter. IPSO declines to act because British press regulation permits fabricated quotes. The procedure exists. The procedure does nothing. Turner’s critique of how rules performing fairness produce unfairness applies. Same with Cambridge’s Free Speech policy. The policy exists. The policy did not save Cofnas’s college position. The university wrote it after Noah Carl to make this case impossible, and the case happened anyway.
The hereditarian content has its own essentialism problem. Cofnas hedges in places. He distinguishes Appalachian Whites from Massachusetts Whites. He distinguishes selected African immigrants in the UK from population-representative African samples. He notes cultural transmission as a force and accepts that genes do nothing alone. These hedges cut against the strong hereditarian frame his audience wants. They also cut against the Cambridge frame, which has its own essentialism, racial categories as fixed identities requiring institutional protection.
The Einstein-Whitten point unsettles both sides. Cofnas grants that Einstein got lucky on historical fit. The same trait, dropped into 1925 instead of 1905, might produce a less famous physicist. If trait plus context determines outcome, then the IQ-predicts-success claim runs into a credence-good problem. We cannot separate the trait from the timing. Hereditarianism explains some of the variance some of the time, mediated by conditions the framework does not control.
The interview enacts coalition discipline. Cofnas criticizes Dawkins for courage on settled questions and silence on live ones. The criticism does coalition work. It says: the real bravery is on our terms, not theirs. Gold endorses this framing. Both speakers position the New Atheist generation as predecessors who went only part of the way. The critique recruits the audience to the next stage of bravery, which is also the next stage of coalition.
Nick Ferrari functions as a third coalition. He is the mainstream anti-woke broadcaster who polices the boundary between acceptable and unacceptable dissent. Cofnas names the move: the mainstream anti-wokers want to prove they are the good ones. Ferrari produces a Black student who feels uncomfortable, the same procedure he claims to oppose. The procedure of the offended bystander floats free of any particular ideology. Any coalition that wants to expel a member without arguing the substance can use it.

Read on.