This site is a Work In Progress

Site Navigation

NeoBooks In Progress

Creating a NeoBook (start here)

Admin and Help

Edit on GitHub


What Went Wrong: Wokeness as a Delivery Failure (Claude analysis)

A sensemaking map — not about whether the underlying concerns were real (they were), but about why the mechanisms chosen to address them so reliably produced the opposite of their stated aims.

The frame

The intuition you started with is the right one: there are at least two separable stories here. One is strategic weaponization — operatives who saw that amplifying the most extreme expressions of progressive moralism would ignite a backlash coalition. That's real, and well-documented, but it's downstream of the second story.

The second story is implementation failure — that the dominant delivery mechanisms of US-style wokeness (mandatory DEI training, diversity bureaucracies, certain framings of affirmative action, slogans like "defund the police," call-out culture, language policing) were designed in ways that predictably backfired on their own terms, even if no political opponent ever exploited them. The strategists had so much to work with because the mechanisms generated their own opposition.

This is the more useful story for sensemaking because it's a story about craft — about what it would mean to actually do this work well. And craft can be learned. Weaponization is just what happens to anything in a contested political environment.

The deepest thing to notice up front: most of these mechanisms were transplanted from elite institutional contexts (universities, HR departments, foundation-funded nonprofits) into mass contexts where their assumptions didn't hold. They assumed a captive audience, a shared moral vocabulary, an acceptance of the diagnostician's authority, and a willingness to perform compliance. None of those hold outside the seminar room.

The failure modes

Each of these is a mechanism — a specific way an intervention turned into its opposite. The research below is mostly from social psychology, organizational behavior, and political communication.

1. Psychological reactance — the door-slam reflex

When people perceive that a freedom is being threatened — including the freedom to form their own opinion on a contested question — they don't just resist the imposition. They often move toward the prohibited position and hold it more strongly than before. Brehm formalized this in the 1960s; it's one of the most replicated findings in social psychology.

Mandatory DEI training is a reactance machine by design. You sit adults down, tell them their existing views are wrong, and require them to affirm a new framework as a condition of employment. The predictable response — not in everyone, but in a substantial fraction — is private hardening of exactly the views the training was meant to soften. The training doesn't fail to change minds; it changes them in the wrong direction.

The same dynamic operates on language. Tell people they may not say a word, and the word acquires a charge it didn't have before. The prohibition becomes the offense.

2. Moral licensing — the credit you gave yourself

When people perform a virtuous act, they subsequently feel licensed to behave less virtuously, as if they'd banked moral credit. Monin and Miller showed this in the early 2000s; it's been replicated across domains from charity to environmental behavior to hiring.

A DEI training session, a land acknowledgment, a posted statement of values, a diversity hire — each can function as a license to stop thinking about the underlying issue. The institution has done the thing. The box is checked. Real changes to how decisions get made, where resources flow, who holds power — those become less likely, not more, because the ritual has already discharged the moral demand.

This is why Dobbin and Kalev found that companies with the most elaborate diversity programs often had worse representation outcomes than companies with simpler, more targeted interventions. The programs were doing the moral work the actual changes were supposed to do.

3. Stereotype threat in reverse — the affirmative action wound

Claude Steele's stereotype threat research showed that members of stigmatized groups underperform when their group identity is made salient in a context where they expect to be judged by it. The original finding is well-established.

A less-discussed corollary: when membership in an advantaged group is made salient as the basis for selection or non-selection, the result is not humility but resentment plus a sense that one's own achievements are being discounted. This is the affirmative action wound on the non-minority side, and it's psychologically real even when the policy is materially justified.

The deeper problem is that demographically-keyed remedies re-essentialize the categories they're trying to dismantle. You can't tell people race is a social construction with no biological basis and then build a permanent administrative apparatus that sorts them by race. One of those two claims has to give, and in practice it's the first one — the categories harden into the bureaucratic infrastructure, which then needs them to continue existing.

4. The backfire of public shaming — call-out culture

Loretta Ross, a Black feminist organizer who helped coin "reproductive justice," became one of the sharpest critics of call-out culture from inside the left. Her argument: public shaming is a tool of the powerless against the powerful, and when it's turned sideways or downward — against allies, against people who made minor errors, against people who can be reformed — it produces fear, silence, and resentment rather than change. The people who could have been brought along instead learn to keep their mouths shut and quietly hate the movement.

Erica Chenoweth's research on nonviolent movements points the same direction from a different angle: movements succeed by expanding their coalition, particularly by peeling off members of the opposition's base. Mechanisms that contract the coalition by demanding ideological purity from anyone who wants to participate are anti-movement by definition.

5. The slogan problem — semantic capture by opponents

"Defund the police" is the case study, but it's a category. A slogan that requires immediate explanation has already failed as a slogan. The moment supporters have to say "well, what we actually mean is..." they've ceded the framing to whoever defines it first, and in a contested information environment that will be the opposition.

Lakoff's frame-semantics work is the obvious reference here. Frames are sticky; once a phrase activates a particular conceptual structure ("abolish the institution that protects me from crime"), the technical policy content is invisible underneath the frame. The intended audience for "defund" was internal — people already in the movement who'd parsed it in policy terms. The actual audience was everyone, and everyone heard what it said.

Compare to how the marriage equality movement chose its frame: "love is love," "marriage equality," not "abolish heteronormative kinship structures." Even people who agreed with the deeper structural critique knew not to lead with it.

6. The diagnostic-without-treatment problem

A pattern across many DEI curricula: spend most of the time getting participants to acknowledge the depth and intractability of systemic bias, then offer either no actionable response or only individual-attitudinal responses ("examine your privilege," "do the work"). This produces what John McWhorter has called a kind of secular original-sin framework — a permanent state of guilt with no possibility of absolution and no clear theory of change.

People can tolerate being told they're complicit in something if there's a path out. They can't tolerate being told they're permanently complicit in something with no path out. The predictable response is to reject the diagnosis altogether.

7. Elite capture and the credentialing of grievance

Olúfẹ́mi Táíwò's Elite Capture makes the argument cleanly: identity-based frameworks designed to surface the voices of marginalized people get captured, in elite institutional settings, by the most credentialed and least marginalized members of those groups. The Yale-educated child of professionals speaks for "the Black experience"; the actually-marginalized person, whose interests might cut against the institutional consensus, is unheard.

This isn't a critique of the underlying frameworks. It's a critique of what happens to them when they become career paths inside wealthy institutions. The DEI officer's job security depends on the persistence of the problem they're hired to address. The incentive structure runs opposite to the stated mission.

8. The collapse of the public/private distinction

Older labor and civil rights traditions distinguished sharply between what could be required in the workplace (non-discrimination, equal treatment, professional conduct) and what was a matter of private conscience (what you believe, what you value, who you vote for). The newer DEI framework tends to collapse this — demanding not just compliant behavior but affirmative endorsement of particular worldviews as a condition of professional participation.

This generates resistance even from people who broadly agree, because it violates a deep liberal intuition about the limits of institutional power over conscience. You can require people to treat colleagues fairly. The moment you require them to believe particular things about why fairness is owed, you've crossed a line that liberalism spent centuries establishing.

What the research actually shows about DEI

Frank Dobbin (Harvard) and Alexandra Kalev (Tel Aviv) have done the most rigorous longitudinal work on this, tracking outcomes across more than 800 US firms over three decades. Their central findings, oversimplified:

The shape of the finding is consistent: interventions that change what gets done tend to work; interventions that try to change what people believe tend to backfire. This is the inverse of where DEI spending has actually flowed.

Thinkers who would have done it differently — a curated mix

I'm pulling from several traditions deliberately, because no single tradition has this fully right.

The pragmatic organizational scholars

Frank Dobbin and Alexandra Kalev — Getting to Diversity (2022) is the empirical anchor. Read this if you read nothing else. Their argument is that we know what works; we've just been doing the other thing because it's easier to sell.

Amy Edmondson — psychological safety research. Her framework starts from the question "what conditions let people speak up, take interpersonal risks, and learn from mistakes?" and treats inclusion as an emergent property of those conditions rather than something achievable by direct mandate. The implementation pathway is leader behavior and team norms, not training modules.

Iris Bohnet — What Works: Gender Equality by Design. Behavioral-economic approach. Don't try to change minds; redesign the decision environment so biased decisions are harder to make. Blind auditions for orchestras is the canonical example.

The heterodox liberals

John McWhorter — Woke Racism is polemical but the underlying linguistic point is sharp: he treats certain strands of antiracism as a religious framework with its own original sin, sacraments, and heresy categories, and argues this makes them resistant to evidence and bad at producing change. Whether or not you accept the religion frame, the analysis of why the movement is structurally unable to update is useful.

Glenn Loury — economist, formerly establishment, now heterodox. His work on the difference between "discrimination in contracts" (what civil rights law addressed) and "discrimination in contacts" (informal social capital, networks, who knows whom) is foundational and largely ignored by mainstream DEI. The mechanisms he points at are the ones that actually transmit advantage across generations.

Yascha Mounk — The Identity Trap makes the case that identity-essentialist frameworks, even with the best intentions, undermine the universalist coalitional politics that have historically delivered progress for marginalized groups.

Jonathan Haidt — The Coddling of the American Mind (with Lukianoff) on how three "great untruths" propagated through universities and into the broader culture: what doesn't kill you makes you weaker, always trust your feelings, life is a battle between good people and evil people. The last one is the most politically corrosive.

The restorative and relational traditions

john powell (lowercase deliberate) — directs the Othering & Belonging Institute at Berkeley. His "targeted universalism" framework is the most sophisticated alternative I know: set universal goals, but design group-specific strategies for getting different populations there based on their actual situations. It dodges both the colorblindness trap and the essentialism trap. If I had to point at one thinker who's thought this through most carefully, it would be him.

bell hooks — Teaching to Transgress, All About Love. Her pedagogical work assumes that liberation work is relational, slow, and requires actual love and curiosity across difference. The opposite of the call-out posture. Worth rereading in the current moment.

adrienne maree brown — Emergent Strategy, and more pointedly We Will Not Cancel Us. From inside the movement, arguing that the culture of disposability and purity-policing is itself a form of harm and reproduces the carceral logic the movement claims to oppose.

Loretta Ross — the call-in alternative to call-out culture. Her premise: people can be reached, even people who say ugly things, and a movement that can't reach people isn't a movement, it's a clubhouse.

The deeper civilizational frames

Albert Hirschman — The Rhetoric of Reaction. Not about wokeness directly, but the single most useful book for understanding how progressive arguments get answered (perversity: it will backfire; futility: it won't work; jeopardy: it'll endanger prior gains) — and crucially, how progressive arguments themselves can fall into mirror-image traps. The honest reader notices their own side using these moves too.

James C. Scott — Seeing Like a State. About how high-modernist projects fail when they try to impose legible schemas on complex social reality. DEI as a managerial framework is a high-modernist project in his sense: it tries to make the moral life of an organization legible to administrators, and in doing so destroys the tacit relational textures that actually carry the work.

Ivan Illich — Tools for Conviviality, Disabling Professions. His argument that professional classes capture domains of human life, render ordinary people incompetent in those domains, and then sell back services to address the incompetence they created — this is almost too on-the-nose for the DEI industry.

Humberto Maturana — since this is in your Brain already: structural coupling and the impossibility of "instructive interaction." You cannot inject content into another autopoietic system and expect it to land as intended; the system responds according to its own structure. Mandatory training assumes instructive interaction. Maturana would have predicted it couldn't work.

A synthesis you might find useful

If you wanted one sentence: the dominant US wokeness mechanisms were designed by and for credentialed institutional actors, assumed instructive interaction would work, ignored what's known about adult learning and behavior change, and generated a backlash coalition by treating the freedom of conscience as an obstacle rather than a constraint.

The alternative isn't to abandon the underlying concerns. It's to take craft seriously: relational pedagogy instead of mandatory training, structural redesign instead of attitudinal moralism, targeted universalism instead of demographic sorting, calling-in instead of calling-out, expanding coalitions instead of policing them, and recognizing that the people you're trying to reach are autopoietic systems with their own structures, not blank slates to write on.

This is, not incidentally, what Design from Trust would have suggested all along.

Threads to pull