# DEI and the Myth of American Meritocracy Armchair Authorities | Political Mythology | Script draft 1 Prepared September 18, 2026. Written for a single narrator, with optional cutaways and on-screen jokes. Allow approximately 25–30 minutes, depending on delivery, pauses, and footage. ## How to use this draft Everything outside square brackets and this opening note is intended as narration until the “Editorial and production notes” section. Links are research annotations, not spoken text; they can become lower-thirds or video-description citations. Bracketed directions are suggestions, not required assets. The argument is that equal opportunity requires actively identifying and removing barriers to talent. It is not that every program called DEI works, that every demographic disparity proves discrimination, or that demographic identity establishes anyone’s qualifications. The political section uses documented 2025 appointments rather than assuming that every official remains in the same position at publication. ## Cold open: The best person for the job `[VISUAL: A blank job application. One field appears: “Can you do the job?” Then a second: “Do we like your last name?”]` Imagine that you are hiring a surgeon. You probably want someone who knows where the organs go. You want training. Experience. Good judgment under pressure. Ideally, somebody whose relationship with the human body extends beyond having one. Now imagine two hiring systems. In the first, you look for the most capable person you can find. In the second, you look for the most capable person you can find, but only among people who went to the right schools, knew the right people, and made it through a selection process you have never checked for bias. Both systems can announce, with complete confidence, “We hire on merit.” Only one has even tried to find out whether that is true. That is the question I want to put at the center of the DEI debate. Not whether qualifications matter. Whether our institutions are actually identifying them. In January 2025, President Trump signed an executive order titled “Ending Illegal Discrimination and Restoring Merit-Based Opportunity,” explicitly presenting illegal DEI preferences as a threat to hard work, excellence, and individual achievement. ([White House executive order](https://www.whitehouse.gov/presidential-actions/2025/01/ending-illegal-discrimination-and-restoring-merit-based-opportunity/)) “Restoring merit” is a powerful promise. But before we restore something, we should probably establish when we had it. `[TITLE CARD: “DEI AND THE MYTH OF AMERICAN MERITOCRACY”]` ## The promise and the mythology Here is the version of meritocracy worth defending: your opportunities should depend on what you can do, not whether you were born into the right family or the right demographic. Work should matter. Skill should matter. Your ability to solve a problem should matter more than your ability to recognize the hiring manager’s golf club. That is an aspiration. It becomes mythology when we quietly replace “this is how society should work” with “this is how society already works.” Because those are completely different claims. If you assume the system already rewards people fairly, then the people at the top must deserve to be there. The people at the bottom must deserve that, too. And any attempt to change who gets through the gate starts to look like interference with a fair competition. But if the gate itself is biased, leaving it alone is not neutrality. It is a decision to preserve the bias. There is a historical irony here: Michael Young’s 1958 book, The Rise of the Meritocracy, popularized the idea through satire, warning about a society whose successful people became convinced that their advantages proved their superiority. ([Michael Young, “Down with meritocracy”](https://www.theguardian.com/politics/2001/jun/29/comment)) Decades later, Young wrote that appointing people on merit made sense, but allowing those judged meritorious to harden into a closed social class did not. ([Michael Young](https://www.theguardian.com/politics/2001/jun/29/comment)) `[VISUAL: Book jacket stamped “WARNING.” The stamp rotates into “MISSION STATEMENT.”]` This is an impressive misunderstanding of a book. It is like reading a fire-safety manual and coming away really enthusiastic about fire. And there is another distinction worth protecting. Someone can deserve a particular job without deserving infinitely more dignity than the person who did not get it. A fair hiring process is not a theory of human worth. ## The competition was not starting from zero To see the problem with treating meritocracy as an accomplished fact, consider the conditions under which people were allowed to develop and demonstrate their abilities. The Civil Rights Act of 1964 prohibited employment discrimination and segregation in public accommodations, confronting exclusion that had been embedded in American institutions. ([National Archives](https://www.archives.gov/milestone-documents/civil-rights-act)) That was not a national intervention against an excess of fairness. Housing provides another example: research on federal mortgage programs documents how the Federal Housing Administration’s early underwriting practices promoted segregation and restricted mortgage insurance in the neighborhoods where many Black Americans lived. ([Fishback and colleagues, “New Evidence on Redlining”](https://www.nber.org/system/files/working_papers/w29244/w29244.pdf)) Now follow the logic forward. If one family can buy an appreciating home and another cannot, those families may have different resources available for their children. If one student can afford an unpaid internship and another needs a paycheck, the same internship requirement does not create the same opportunity. If one applicant has been introduced to the industry since childhood and another has never met anyone in it, “just use your network” is not neutral advice. It is an instruction to produce an asset they may never have had the chance to acquire. `[VISUAL: Two applicants receive the same instruction: “Use your connections.” One opens a contact list; the other opens an empty spreadsheet.]` None of this means the first applicant is untalented. None of it means the second applicant is automatically better. It means the résumé is a record of opportunities as well as abilities. And this is where the debate gets needlessly personal. Saying that a barrier advantages one group does not mean every member of that group has an easy life. A struggling white man is not secretly living in a country club because somebody used the word “privilege.” The useful question is specific: which obstacles did this person face, which advantages did this person receive, and which of those actually tell us whether they can do the work? You can recognize racial discrimination and class disadvantage at the same time. You do not have to pick one injustice and pretend the others were discontinued. ## What DEI actually means Let’s take the acronym apart without turning this into the training module you are trying to finish before lunch. First, an important qualification: DEI is a broad label, not a single policy, and even the EEOC notes that it is not a term defined in Title VII. ([EEOC guidance](https://www.eeoc.gov/wysk/what-you-should-know-about-dei-related-discrimination-work)) For this discussion, here is the practical framework. Diversity asks who is present and whose experience is represented. Equity asks whether people have a genuinely fair opportunity, including whether unnecessary barriers are blocking them. Inclusion asks whether people can actually participate and contribute once they arrive. `[ON SCREEN: “Who gets considered? / What blocks access? / Who gets to contribute?”]` Those are working definitions, not a promise that every organization uses the words consistently. Imagine an employer that recruits exclusively through the founder’s friends. Expanding recruitment does not require hiring somebody unqualified. It means the founder’s social circle is no longer being mistaken for the entire supply of human talent. Imagine a job interview where one candidate gets technical questions and another gets an hour of friendly conversation about a shared hobby. Standardizing the questions does not lower the bar. It makes the bar less dependent on whether the interviewer wants to have a beer with you. In fact, the Office of Personnel Management describes structured interviews as using the same predetermined questions and common rating standards to assess job-related competencies consistently. ([OPM structured-interview guidance](https://www.opm.gov/policy-data-oversight/assessment-and-selection/structured-interviews/)) That is a process reform. You can use it without calling it DEI at all. Now imagine an office that hires a wheelchair user but holds important meetings in an inaccessible space. Getting someone through recruitment is not the same as enabling them to do their job. For our purposes, equity is about a fair chance to compete, not a guarantee that everyone wins. Inclusion is about participation, not immunity from criticism or performance standards. But we should also avoid winning this argument by definition. If a program treats demographic identity as a substitute for competence, describing it as “equity” does not resolve the objection. If a diversity target becomes a rigid instruction to disregard qualifications, it deserves scrutiny. Title VII’s protections apply to workers of every race and sex, and a DEI label does not exempt an employment decision from discrimination law. ([EEOC guidance](https://www.eeoc.gov/wysk/what-you-should-know-about-dei-related-discrimination-work)) The standard should be simple: defend fair access, evaluate the actual policy, and do not give your preferred acronym diplomatic immunity. ## Why “just hire the best” is not enough The strongest evidence that this problem is real does not come from asking employers whether they consider themselves fair. It comes from changing what an employer sees while controlling the qualifications. In a well-known field experiment published in 2004, Marianne Bertrand and Sendhil Mullainathan sent fictitious résumés to employers in Boston and Chicago, randomly assigning names intended to signal race. ([American Economic Review](https://www.aeaweb.org/articles?id=10.1257/0002828042002561)) Résumés assigned white-sounding names received about 50 percent more interview callbacks than those assigned Black-sounding names. ([American Economic Review](https://www.aeaweb.org/articles?id=10.1257/0002828042002561)) `[ON SCREEN: “50% more callbacks. Not 50 percentage points. Not final hiring decisions.”]` The name was not a job skill. And this is not only a story about newspaper job ads from an earlier era. Patrick Kline, Evan Rose, and Christopher Walters later studied more than 83,000 fictitious applications submitted to jobs at 108 large U.S. employers. ([NBER study, subsequently published in the Quarterly Journal of Economics](https://www.nber.org/papers/w29053)) Applications with distinctively Black names had a contact probability 2.1 percentage points lower than applications with distinctively white names. ([Kline, Rose, and Walters](https://www.nber.org/papers/w29053)) Importantly, the discrimination was not evenly distributed: the firms in the highest-discrimination fifth accounted for nearly half of the lost contacts to Black applicants in the experiment. ([Kline, Rose, and Walters](https://www.nber.org/papers/w29053)) That variation matters. We do not need to pretend every employer behaves identically to recognize a serious problem. These experiments measure access to the next stage, not the eventual performance of every applicant. They do not establish that every rejected person was the best candidate. ([Bertrand and Mullainathan](https://www.aeaweb.org/articles?id=10.1257/0002828042002561); [Kline, Rose, and Walters](https://www.nber.org/papers/w29053)) What they do undermine is the comforting assumption that qualifications are the only thing being evaluated. “Just hire the best person” is an excellent goal. It is not, by itself, a method. You still need to ask whether you found that person, whether you assessed them accurately, and whether irrelevant information changed your judgment. Otherwise, “merit” can become whatever your institution already happens to reward. `[VISUAL: A hiring rubric containing one criterion: “Reminds us of ourselves.”]` ## The talent we were leaving on the table There is a larger consequence to excluding capable people: society loses the work they could have done. A 2019 paper by Chang-Tai Hsieh, Erik Hurst, Charles Jones, and Peter Klenow examined changes in the allocation of talent across occupations between 1960 and 2010. ([Stanford summary of the Econometrica paper](https://www.gsb.stanford.edu/faculty-research/publications/allocation-talent-us-economic-growth)) One striking starting point: in 1960, 94 percent of doctors and lawyers were white men; by 2010, that share was 62 percent. ([Hsieh and colleagues](https://www.gsb.stanford.edu/faculty-research/publications/allocation-talent-us-economic-growth)) You could explain that earlier concentration by assuming almost all the relevant talent happened to be born into one demographic. Or you could investigate the barriers that kept other people from developing and using theirs. Using an economic model, the researchers estimated that improved talent allocation could explain between 20 and 40 percent of the growth in market output per person over that fifty-year period, depending on the specification. ([Hsieh and colleagues](https://www.gsb.stanford.edu/faculty-research/publications/allocation-talent-us-economic-growth)) `[ON SCREEN: “20–40% of growth in market output per person, 1960–2010. Model-based estimate.”]` That is not “DEI created 40 percent of the economy.” It is not a randomized experiment on corporate diversity departments. The paper studies a much broader historical change and estimates its contribution through a model. ([Hsieh and colleagues](https://www.gsb.stanford.edu/faculty-research/publications/allocation-talent-us-economic-growth)) But the mechanism is the heart of this argument. If a talented potential doctor cannot become a doctor, the loss is not confined to that person. It also belongs to the patients they might have treated. If a capable engineer is screened out for an irrelevant reason, protecting that barrier is not protecting engineering standards. We should not describe the discovery of previously excluded talent as a sacrifice of merit. It can be an improvement in our ability to recognize merit. Now, does that mean every individual benefits equally from every change? No. If a competition becomes more open, somebody who previously faced less competition may lose an advantage. That can feel like a loss. But losing an advantage over other candidates is not automatically the same as being treated unfairly. If I only win the audition when half the musicians are kept outside, letting them audition has not destroyed musical excellence. It has created an uncomfortable afternoon for me. `[VISUAL: Narrator holding a guitar. Off-screen, someone plays a much better riff.]` I would prefer that we not test this analogy personally. ## What organizations actually did, and what worked So far, we have established a problem and a reason to address it. We have not established that everything sold as the solution works. That distinction matters. Research on organizational diversity efforts has examined a range of actual practices, including affirmative-action plans, diversity managers, committees, training, networking, and mentoring. ([Kalev, Dobbin, and Kelly, 2006](https://journals.sagepub.com/doi/abs/10.1177/000312240607100404)) Even historically, federal-contractor affirmative action was not formally synonymous with an instruction to hire unqualified people: a Clinton-era government review distinguished flexible goals and good-faith efforts from prohibited quotas, while also acknowledging complaints about rigid implementation. ([Archived White House affirmative-action review](https://clintonwhitehouse3.archives.gov/WH/EOP/OP/html/aa/aa06.html)) That is a useful distinction, not a guarantee that every employer followed the rules. And that federal framework changed: Trump’s January 2025 order revoked Executive Order 11246, the longstanding foundation for race- and sex-related affirmative-action requirements for federal contractors. ([White House executive order](https://www.whitehouse.gov/presidential-actions/2025/01/ending-illegal-discrimination-and-restoring-merit-based-opportunity/)) But whether we are discussing a government requirement or a voluntary company initiative, the evaluation question should be the same: what changed in practice? Alexandra Kalev, Frank Dobbin, and Erin Kelly studied 708 private-sector establishments using workforce data from 1971 through 2002 alongside information about employment practices. ([Kalev, Dobbin, and Kelly](https://journals.sagepub.com/doi/abs/10.1177/000312240607100404)) They found that practices establishing organizational responsibility for diversity were associated with the broadest increases in managerial diversity; training and diversity evaluations aimed at changing managerial bias were the least effective approaches on that measure. ([Kalev, Dobbin, and Kelly](https://journals.sagepub.com/doi/abs/10.1177/000312240607100404)) Notice the outcome: representation in management. Not proof that every decision became fair, and not proof of a universal increase in profits. The study was also not a randomized assignment of companies to programs. Treat its findings as evidence about organizational patterns, not a guarantee that installing a committee will fix your workplace. There is another caution from psychology. A 2019 meta-analysis covering 492 studies found that interventions could change implicit measures, but changes in those measures did not reliably translate into changes in behavior. ([Forscher and colleagues](https://pmc.ncbi.nlm.nih.gov/articles/PMC6687518/)) Much of that evidence involved brief interventions and short-term measurement, so it is not a verdict on every possible long-term training program. ([Forscher and colleagues](https://pmc.ncbi.nlm.nih.gov/articles/PMC6687518/)) Still, it is a very good reason not to confuse “everyone completed the workshop” with “we solved discrimination.” `[VISUAL: Certificate reading “Congratulations! You have successfully clicked Next.”]` A company can measure attendance because attendance is easy to measure. That does not make attendance the outcome we care about. The broader research on workplace diversity also rejects the idea of an automatic benefit: effects on performance, integration, and well-being depend on organizational conditions and how differences are managed. ([Guillaume and colleagues’ review](https://pubmed.ncbi.nlm.nih.gov/28239234/)) So the honest answer to “Does DEI work?” is: which intervention, in what setting, measured against what outcome? That is not dodging the question. That is finally asking a question precise enough to answer. ## What DEI did not prove, and what fairness would require Here are two arguments I do not think we should make. The first is that any demographic imbalance proves discrimination. A disparity can be a reason to investigate; it is not, by itself, a complete explanation. The second is that any policy producing more demographic diversity must therefore be fair or effective. You still have to examine the process, the relevant qualifications, and the consequences. But the reverse shortcut is just as bad: seeing a woman or a person of color in a role and treating their presence as evidence that the standards were lowered. You cannot infer the hiring process from someone’s face. And you cannot establish that a diversity initiative caused a failure merely by pointing out that an organization had one. You need evidence connecting a specific policy to the specific decision and the resulting harm. By the same logic, we cannot say that a successful organization proves all of its diversity policies worked. Causation should not change its requirements depending on whose argument it inconveniences. What would a more credible approach look like? Start by defining the job. Identify which skills actually matter. Publish the opportunity broadly. Assess candidates against consistent, relevant criteria. Investigate unexplained differences rather than inventing an explanation in advance. Then examine development and retention, not just the first hiring decision. If important opportunities are distributed informally, ask whether everyone who could benefit has a fair chance to access them. And keep measuring performance. Fair access and high standards should reinforce each other. This can also make room for people who do not fit the usual political picture of a DEI beneficiary. Consider a rural white applicant shut out by geography, a veteran whose experience is poorly translated into civilian job requirements, or a working-class student unable to afford an unpaid placement. Those are hypothetical examples, not proof that every DEI program serves those populations. The principle is that a fair system should identify the barrier rather than decide in advance who is allowed to have one. It is also worth noting that Trump’s own January 2025 order explicitly exempted lawful employment and contracting preferences for veterans. ([White House executive order](https://www.whitehouse.gov/presidential-actions/2025/01/ending-illegal-discrimination-and-restoring-merit-based-opportunity/)) You can support that exception. But it should make the conversation more thoughtful than “any consideration beyond a context-free ranking is automatically an attack on merit.” ## The administration’s own merit test `[VISUAL: The executive-order title returns. Underline “Merit-Based Opportunity.”]` Now let’s apply that standard upward. If your political argument is that institutions must stop elevating people for reasons unrelated to competence, your own appointments are a reasonable place to look. Take Pete Hegseth’s selection to lead the Pentagon in 2025. Hegseth was a decorated military veteran and former Fox News host, so dismissing him as someone with literally no relevant experience would be inaccurate. ([Reuters, January 2025 confirmation reporting](https://www.reuters.com/world/us/hegseth-narrowly-wins-confirmation-become-us-defense-secretary-2025-01-25/)) But Reuters reported that he told lawmakers the largest group he had managed was 100 people and the largest budget was $16 million, before taking responsibility for an institution employing millions and operating on a vastly larger budget. ([Reuters](https://www.reuters.com/world/us/hegseth-narrowly-wins-confirmation-become-us-defense-secretary-2025-01-25/)) Compare that preparation with his predecessor, Lloyd Austin, whose 41-year Army career included commanding U.S. forces in Iraq, serving as Army vice chief of staff, and leading U.S. Central Command. ([Official Austin biography](https://www.war.gov/About/Biographies/Biography/article/2522687/lloyd-j-austin-iii/)) That comparison does not prove Austin made every decision correctly. Experience is evidence of preparation, not a certificate of infallibility. Nor should we require every defense secretary to have the same military résumé. The job calls for civilian oversight, and there can be more than one route to competence. But it is entirely reasonable to ask why a movement so concerned about standards was comfortable with that managerial experience gap. `[ON SCREEN: “TRANSFERABLE SKILLS: please elaborate.”]` Then consider Robert F. Kennedy Jr., confirmed as health secretary in February 2025 despite substantial opposition over his anti-vaccine advocacy. ([Reuters, February 2025](https://www.reuters.com/world/us/kennedy-clear-final-senate-hurdle-bid-top-us-health-job-2025-02-13/)) The useful criticism is not simply that he was a lawyer rather than a physician. His predecessor, Xavier Becerra, was also a lawyer, with prior experience in Congress and as California attorney general. ([Reuters on Kennedy](https://www.reuters.com/world/us/kennedy-clear-final-senate-hurdle-bid-top-us-health-job-2025-02-13/); [Becerra’s congressional biography](https://www.congress.gov/118/meeting/house/116999/witnesses/HHRG-118-WM00-Bio-BecerraX-20240320.pdf)) The question is judgment: how should a public-health leader evaluate evidence, and what does their record tell us about how they will do that? At Kennedy’s confirmation, Reuters documented his promotion of a discredited vaccine-autism link, alongside his promises to protect vaccination programs and work within existing safety systems. ([Reuters](https://www.reuters.com/world/us/kennedy-clear-final-senate-hurdle-bid-top-us-health-job-2025-02-13/)) His supporters presented him as a challenger to entrenched interests and a champion against chronic disease; those objectives do not, by themselves, answer the concerns about his scientific judgment. ([Reuters](https://www.reuters.com/world/us/kennedy-clear-final-senate-hurdle-bid-top-us-health-job-2025-02-13/)) And the political route to his nomination was not mysterious: Reuters reported that he ended his independent presidential campaign and endorsed Trump in exchange for a role in the administration. ([Reuters, November 2024](https://www.reuters.com/world/us/trump-expected-select-rfk-jr-lead-hhs-politico-reports-2024-11-14/)) Political alignment is not inherently an illegitimate consideration in a political appointment. But it is not the same thing as demonstrated competence. So here is the satirical reversal: if you redefine “DEI” to mean “giving an important job to someone because they belong to your preferred group rather than because they have demonstrated the strongest relevant qualifications,” then political in-group hiring deserves the same suspicion. Call it D.E.I.: Donald’s Employment Initiative. `[BEAT. ON SCREEN: “Satirical acronym. Not an actual program.”]` The serious term is favoritism or patronage, where the evidence supports it. That is not what diversity means, and I am not endorsing the idea that diverse hires are inherently unqualified. I am asking for a consistent standard. If you want evidence of competence from somebody else’s appointee, ask for it from your own. If an outsider deserves a chance to prove themselves, explain why that principle disappears when the outsider belongs to a demographic you dislike. Merit cannot mean “scrutiny for them, benefit of the doubt for us.” ## Closing: Take merit seriously The case for fair opportunity does not require pretending corporate DEI was a flawless machine. We should be willing to criticize a bad workshop, an unfair selection rule, a hollow publicity campaign, or a program that cannot demonstrate what it accomplished. None of those deserves protection from evaluation. But an ineffective remedy does not establish that the underlying problem was imaginary. And abolishing a department does not, by itself, produce an unbiased hiring process. The principle worth keeping is bigger than the acronym. If merit matters, we should care whether people can develop their abilities. We should care whether capable applicants are being overlooked. We should care whether the criteria actually measure the work. We should be just as willing to challenge an exclusionary old network as an unfair new initiative. Because the choice is not between merit and inclusion. The choice is between assuming we have a fair competition and doing the work required to create one. Nobody should get a job merely because of their identity. Nobody should lose the chance to earn it because of their identity, either. That is the standard. [VISUAL: Return to the original application. The irrelevant fields disappear. “Can you do the job?” becomes “Show us what you can do.”] You do not defend meritocracy by declaring the winners meritorious. You defend it by making sure talent has a fair chance to be found. This is Armchair Authorities. If you disagree, tell us which policy or claim you would change, and why. The research is linked so you can check the argument, not just borrow our confidence. ## Editorial and production notes These notes are not narration. They document the boundaries of the argument and the main choices for the next revision. - **Core thesis:** Keep “removing barriers is necessary for meaningful merit-based opportunity” distinct from “every existing DEI program is necessary.” The first is the argument; the second is not established by this evidence. - **Growth estimate:** The 20–40% figure concerns the share of growth in market output per person over 1960–2010 explained by improved talent allocation under alternative model specifications, not a measured effect of contemporary DEI departments or a sudden 1970s spike. ([Published-paper summary](https://www.gsb.stanford.edu/faculty-research/publications/allocation-talent-us-economic-growth)) - **Wage figures in the project notes:** Omitted from this draft rather than presenting potentially model-specific decomposition results as simple observed wage changes. Before adding them, verify the exact table, denominator, counterfactual, and whether each figure describes an observed change or an estimated effect. - **Callback numbers:** Keep relative percentages separate from percentage points, and callbacks or employer contacts separate from job offers. These studies do not identify the eventual best performer in each applicant pool. ([Bertrand and Mullainathan](https://www.aeaweb.org/articles?id=10.1257/0002828042002561); [Kline, Rose, and Walters](https://www.nber.org/papers/w29053)) - **Program efficacy:** Managerial representation, implicit-measure change, workplace behavior, and organizational performance are different outcomes. Do not edit their findings into a blanket “DEI works” or “DEI does not work” statement. ([Kalev and colleagues](https://journals.sagepub.com/doi/abs/10.1177/000312240607100404); [Forscher and colleagues](https://pmc.ncbi.nlm.nih.gov/articles/PMC6687518/); [Guillaume and colleagues](https://pubmed.ncbi.nlm.nih.gov/28239234/)) - **Administration comparison:** “Most unqualified administration in history” would require a defined metric and systematic historical comparison that this draft does not supply. Keep the satire, documented comparisons, and criticism of inconsistent standards; avoid presenting a historical ranking as a researched conclusion. - **Political footage:** Use the executive-order title, appointment coverage, relevant hearing clips, and résumé comparisons. Verify exact transcript wording and editing context before inserting any quoted clip; do not substitute a paraphrase in quotation marks. - **Historical imagery:** Pair the FHA discussion with FHA underwriting material, not an unlabeled HOLC map that implies the two agencies’ practices were identical. The cited research specifically distinguishes them. ([Fishback and colleagues](https://www.nber.org/system/files/working_papers/w29244/w29244.pdf)) - **Legal scope:** This is an essay about policy and evidence, not a comprehensive guide to current employment law. Recheck legal developments before recording; do not broaden the historical discussion of federal-contractor rules into a claim that every affirmative-action policy in every setting used the same rules. - **Release-date check:** The political examples intentionally concern appointments in 2025. If revised to say “today,” check current officeholders, subsequent performance evidence, and the status of relevant executive orders immediately before publication. - **Narration pass:** Read aloud once before locking the edit. Retain enough spoken qualifications to preserve the argument, but move study details and dates into on-screen annotations where they interrupt the delivery.