Key Takeaways
Judge the incident from evidence, not from the identities in the headline
Cases need context. Huemer revisits high-profile shootings and argues that viral accounts often omitted important evidence. In his reading, evidence supports self-defense claims in the Trayvon Martin and Michael Brown cases, while Jacob Blake’s arrest warrant, knife, and resistance change how his shooting video should be understood. He also argues that Kyle Rittenhouse acted in self-defense during the Kenosha shootings.
No blanket verdicts. Huemer also identifies Eric Garner, Breonna Taylor, and George Floyd as wrongful police killings. His recurring point is that neither a badge nor a racial identity settles whether force was justified. He examines witness accounts, forensic evidence, trial testimony, and circumstances in each case, rather than treating a person’s identity as a verdict.
The case-by-case approach is a useful antidote to snap judgment, especially when a short video or headline leaves out events that happened before recording began. It also applies in both directions: evidence that complicates one activist account does not establish that police conduct is generally justified. These cases are vivid, but selected examples cannot establish how common a pattern is. They are best treated as demonstrations of why careful investigation matters, not as a substitute for representative data.
Check the denominator before calling a disparity proof of bias
Counts need context. Huemer notes that many people greatly overestimated police killings of unarmed Black people. For 2019, he cites 36 unarmed Black people killed by police, compared with 54 unarmed white people. He argues that overall racial disparities in police shootings should be compared with relevant factors such as police contacts and rates of violent behavior, not just each group’s share of the population.
Experiments add another angle. In a police simulator study he discusses, officers were slower to shoot armed Black suspects than armed white suspects, and more often mistakenly shot white suspects holding harmless objects. Huemer takes this as evidence against the claim that lethal police shootings broadly reflect anti-Black bias. He does not argue that police use of force is never excessive.
A disparity is a prompt for causal investigation, not a conclusion by itself. Choosing a denominator can change the interpretation, but adjustment also requires care: police contact rates may themselves reflect unequal treatment, and crime statistics are not a neutral measure of every relevant encounter. The simulator helps isolate race in a controlled scenario, while its artificial setting limits how directly its results transfer to street encounters. Good analysis makes both the comparison and its limits visible.
Interrogate what the headline’s statistic actually counts
The pay gap is aggregate. The commonly cited figure that women earn about 82 cents per dollar compares all men and women, not people doing identical work. Huemer reports that the gap shrinks to about one cent after controlling for factors such as occupation, experience, and hours worked. He argues that remaining differences may also reflect job conditions and personal tradeoffs.
Definitions reshape claims. The book challenges the claim that one in four college women are raped, arguing that an extrapolation and broad survey definitions inflate the figure; Huemer estimates completed rape at roughly 1% to 5%. A separate survey’s 26% sexual-assault figure includes a wider range of unwanted contact. Likewise, the often-cited 2% to 10% false-report estimate counts cases proven false, not every accusation that was actually false. The true rate is unknown.
The strongest lesson is about measurement: a statistic inherits the choices used to define, sample, and classify its events. That scrutiny should be applied to every side’s preferred numbers. Huemer’s alternative estimates also depend on assumptions about time periods, survey responses, and what counts as an event; they are not direct measurements of a definitive rate. In charged debates, clear definitions can reduce confusion, but they cannot by themselves resolve questions about harm, responsibility, or fair policy.
Keep sex, gender identity, and social roles conceptually distinct
Terms do different work. Huemer rejects both the claim that gender is only a social role and the claim that it is merely another word for biological sex. He proposes that gender consists of psychological traits associated with a sex, including comfort with sex-typical bodily features and social identification. This is his theory, not a settled definition.
Cases complicate simple rules. David Reimer was born male, assigned female after a medical injury, and later lived as a man. Huemer treats this case, along with some intersex and transgender experiences, as evidence that identity is not always determined by upbringing or appearance. He distinguishes early-onset gender dysphoria from later identification, and notes that some people desist or detransition. He also discusses peer influence as one proposed explanation for some recent adolescent cases.
Separating biological traits, inner experience, and social categories can clarify disagreements that otherwise collapse into competing definitions of “woman” or “man.” Yet the evidence described does not establish one complete theory of gender. A single case history cannot identify the cause of a broad phenomenon, and parent reports about adolescents cannot by themselves distinguish social influence from improved recognition or other explanations. The careful move is to preserve conceptual distinctions while treating causal claims, especially about young people, as questions requiring stronger evidence.
Demand replication before turning a lab effect into public policy
A test is not a diagnosis. The Implicit Association Test measures how quickly people sort paired ideas. Huemer cites low test-retest reliability for its race version, around 0.33, and a small relationship with observed discriminatory behavior. A large review found that changing IAT scores did not reliably change people’s behavior.
Replicate before generalizing. Huemer also challenges stereotype threat, the proposed effect in which awareness of a negative group stereotype lowers performance. He argues that a famous early experiment was commonly misread, that many later attempts failed to reproduce key results, and that effects were negligible in studies using real exams. He places both debates within psychology’s broader replication crisis, where published effects have often proved smaller or failed to reappear in repeat studies.
Replication is not a demand that every study produce identical results. It asks whether an effect survives new samples, clearer methods, and realistic settings. The distinction between measuring an association and predicting an individual’s conduct is especially important for workplace or legal decisions. At the same time, weak predictive power for one test does not prove that bias never operates, and failed replications do not show that every result in social psychology is false. They justify narrower claims and better evidence.
Separate climate findings from catastrophe rhetoric
The warming claim is not the apocalypse claim. Huemer accepts that the planet has warmed by about 1.1°C since the Industrial Revolution and that human activity is probably a major cause. He argues that the famous 97% figure describes papers that expressed a view and supported some human contribution, not 97% of scientists agreeing that humans are the main cause. A survey of meteorologists found less agreement than that headline suggests, while a later survey of geoscientists found more.
Deadlines were distorted. The IPCC’s 2030 target concerned limiting warming to 1.5°C, Huemer explains, not a prediction that civilization would end if the target were missed. He urges attention to projected harms and the cost-effectiveness of proposed responses, rather than treating climate change as either imaginary or inevitably apocalyptic.
Scientific consensus is useful evidence for readers who cannot independently evaluate a technical field, but the quality of that evidence depends on how consensus is measured and how open the field is to criticism. The book usefully separates warming, human contribution, projected harm, and policy choice, claims that are often bundled together. Its skepticism about consensus should not be mistaken for proof that experts are wrong. Conversely, agreement about a physical trend does not automatically settle which interventions are worth their cost.
Treat mask mandates as risk-benefit decisions, not loyalty tests
The evidence is mixed. Huemer reviews studies of face coverings against respiratory viruses, including a 2023 meta-analysis of 78 randomized trials, 12 of which examined surgical masks. The pooled results showed no clear reduction in infection. One large study in Bangladesh found a small difference in reported COVID symptoms between villages receiving a mask-promotion program and comparison villages, but the wider evidence did not establish a large, reliable effect.
Confidence outran results. Early public-health guidance discouraged general mask use, then later officials strongly promoted it. Huemer concludes that masks may offer small benefits, with results depending on mask type and use, and that the evidence did not justify claims that masks were the most powerful protection available. He also discusses possible harms from poor hygiene and risk compensation.
Public-health choices often have to be made before evidence is conclusive, and randomized trials of community behavior are difficult to interpret because adherence varies. A pooled result showing no clear benefit is not the same as proof of zero effect in every setting. Still, officials weaken trust when they present uncertain evidence as certainty or explain policy changes poorly. Transparent communication should distinguish what is known, what remains uncertain, and why a precaution is being recommended.
Separate unequal outcomes from proof of racist intent
The crack-cocaine example. The 1986 federal law set a 100-to-1 quantity ratio between crack and powder cocaine for mandatory penalties. Huemer notes that average sentences shortly before reform were closer to a 1.3-to-1 difference. He also points to differences in the drug markets and to bipartisan support for the law, including support from many Black legislators, as reasons not to treat the disparity alone as proof of a racist legislative purpose.
Intent is not the whole story. Huemer acknowledges that the drug war caused serious harms, including incarceration and damage to communities, and that racism may have played some role in its history. His narrower argument is that evidence of disparate impact does not by itself establish a deliberate racial scheme. The policies may have been supported by lawmakers who believed they would reduce drug-related harm.
Distinguishing intent from impact prevents a common inference error, but neither category cancels the other. A policy can be adopted for non-racist reasons and still impose unequal burdens that deserve correction. Voting records complicate a simple account of who supported the drug war, yet they cannot reveal every legislator’s motives or measure every consequence. Sound policy analysis asks both how a rule arose and whom it harms, then evaluates whether less damaging alternatives could achieve its stated aims.
Check inheritance and mobility before assuming wealth is inherited
Most millionaires did not inherit fortunes. Huemer cites a survey finding that 79% of U.S. millionaires had received no inheritance, and only 3% had inherited more than $1 million. A Treasury study found that 58% of people who began in the lowest income quintile in 1996 had moved to a higher quintile by 2005.
Mobility has multiple causes. Income still correlates with parents’ income, but Huemer argues that the relationship is smaller than many people assume. He cites adoption and twin studies suggesting that inherited traits explain more of the correlation than shared upbringing does. These findings challenge the idea that family wealth is the main route to success, though they do not show that everyone has equal opportunity or that inherited advantages are morally unimportant.
The distinction between inherited money and inherited traits makes the mobility debate more complex than a simple story of either merit or privilege. Even if family income mainly tracks traits rather than parental investment, the resulting inequalities can still raise questions about fairness and public policy. Adoption and twin studies help separate causes, but findings from particular populations do not settle how schools, neighborhoods, discrimination, or luck shape opportunity across the country. Mobility statistics describe movement, not whether the starting line is just.
Compare net tax burdens before demanding the rich pay more
Count taxes and transfers together. Huemer challenges claims that the rich generally pay lower effective taxes. Using Congressional Budget Office data for 2019, he calculates that the top income quintile paid 91% of net federal taxes, while the bottom two quintiles received more in federal transfers than they paid. The top 1% alone accounted for about 34% of the net total.
Watch what counts as income. Warren Buffett’s comparison of his tax rate with his staff’s omitted corporate taxes paid on profits underlying his investments, Huemer argues. Claims that billionaires paid only 3% often compare income taxes with increases in the market value of assets they did not sell. Such gains are unrealized, not cash received. The book’s conclusion is that “fair share” is a moral judgment, but some popular comparisons misstate the tax arithmetic.
Tax debates mix accounting questions with ethical ones. Measuring net taxes gives a fuller picture of transfers, while unrealized gains raise genuine questions about when wealth should be taxed and how to avoid forcing asset sales. The CBO figures cited here cover federal taxes, not every state and local levy, and they do not determine what distribution is fair. Clear accounting can correct misleading comparisons, but it cannot settle the underlying disagreement about redistribution.
Ask who benefits when regulation raises the cost of entry
Rules can serve incumbents. Huemer accepts that some regulations address real harms, but argues that others protect established businesses from competition. Occupational licensing has expanded from about one in twenty U.S. jobs in the mid-twentieth century to about one in four. Licensing can raise prices by making it harder to enter a profession, while evidence that it consistently improves service quality is mixed.
Compliance costs can entrench large firms. A fixed paperwork burden weighs more heavily on a small company than on a major corporation. Huemer argues that large firms may support rules that reduce competition, and that regulators’ close ties to an industry create opportunities for regulatory capture. He also questions reports of high regulatory benefits when agencies themselves supplied the estimates rather than independent reviews of real-world outcomes.
The book’s strongest insight is that regulation has political incentives as well as public purposes. A rule may sound protective while quietly raising barriers that established firms can absorb more easily than entrants. That possibility does not make regulation inherently harmful: licensing can address information gaps or safety risks, and weak competition can create its own abuses. The relevant test is empirical and specific: whether a rule improves outcomes enough to justify its costs, and whether less restrictive options would work.
Verify before sharing, because myths corrode both policy and trust
Beliefs serve more than truth. Huemer argues that political convictions can provide identity, community, and moral status, while social media rewards content that provokes emotion and engagement. He calls the protective arguments that shield an ideology from criticism an “ideological defense system.” Together, these forces help misleading claims spread and make counterevidence easy to dismiss.
Make verification practical. Before repeating a political claim, consult opposing arguments and, where possible, primary sources such as court records, government data, and original research. Look for later studies or meta-analyses rather than relying on a single headline. Huemer warns that myths can produce failed policies, deepen polarization, and teach the public to distrust institutions. His advice is skeptical restraint: when a claim has not been checked, do not amplify it as fact.
This is a valuable standard for political communication because repeating a claim is itself an action, not a neutral gesture. Still, skepticism needs symmetry: if suspicion is applied only to one political camp, it becomes another form of motivated reasoning. Primary sources can be difficult to interpret, and official records are not infallible, so triangulation matters. The deeper civic risk is a cycle in which institutions exaggerate, audiences lose trust, and even accurate warnings become harder to hear.
Analysis
Progressive Myths is an argumentative tour through disputed claims about race, gender, economics, and science, followed by a theory of why political misinformation spreads. Its most useful recurring move is to separate questions that public debate often fuses: disparity from discrimination, social impact from intent, a study’s result from its popular retelling, and an alarming possibility from a likely outcome. The case studies give this method narrative force, while the later chapters explain how group identity, media incentives, and institutional conformity can reward certainty over correction.
The book is strongest when it identifies a concrete claim and examines its underlying measurement or evidence. The pay-gap discussion distinguishes aggregate earnings from comparable work; the climate chapter carefully separates warming from extinction predictions; and the mask discussion considers pooled trial evidence rather than relying on one celebrated study. These examples make an important general point: a claim can be misleading even when the underlying phenomenon is real.
Its central limitation is asymmetry. The book selects progressive claims for scrutiny and often uses that selection to cast doubt on progressive ideology as a whole. Debunking a prominent myth can correct public understanding, but it does not establish that the surrounding political program is generally mistaken. The same caution applies to some of Huemer’s own inferences. Evidence that complicates a racial-bias explanation may weaken that explanation without proving the absence of bias; a study with limitations may warrant uncertainty rather than reversal. Several themes, including the causes of gender identity and the fairness of taxation, also involve contested interpretations that the cited empirical findings cannot settle alone.
The best way to use the book is therefore neither as a replacement political creed nor as a neutral survey, but as a demanding prompt to inspect claims before adopting them. Its durable contribution is methodological: political arguments should remain answerable to evidence, including evidence that unsettles the reader’s preferred story.
Glossary
Progressive myth
Misleading political factual claimHuemer uses this term for an empirical claim believed by many progressives that appears to strongly support progressive ideology but is, in his assessment, false or seriously misleading. The book examines specific examples rather than attempting to refute every progressive position.
Motivated reasoning
Reasoning steered by desired beliefsA way of thinking in which people’s conclusions are influenced by emotional, social, or identity-related preferences, not only by evidence. In the book, political beliefs can help people express group loyalty or a valued self-image, making them more likely to accept congenial claims and resist inconvenient ones.
Implicit Association Test
Timed association measureA psychological test that measures how quickly a person sorts ideas or images into paired categories. In racial versions, participants may sort images of Black or white people alongside positive or negative words. Huemer questions whether its scores reliably measure a stable prejudice or predict discriminatory behavior.
Stereotype threat
Proposed stereotype-related performance effectThe theory that awareness of a negative stereotype about one’s group can create pressure or distraction that lowers performance in a relevant task. The book reviews claims that this explains academic gaps and challenges the size, replicability, and real-world importance of the effect.
Replication Crisis
Failure to reproduce research findingsA broad problem in which findings reported in published studies do not consistently reappear when researchers repeat the studies. Huemer connects the problem to practices such as selective publication and flexible analysis, and argues that claims from psychology should be treated cautiously when they lack robust replication.
Desistance
Return to natal-sex identificationIn the book’s discussion of gender dysphoria, desistance means that someone who previously experienced gender dysphoria later identifies with their natal sex. Huemer distinguishes it from detransitioning, which involves first transitioning socially or medically and later transitioning back.
Autogynephilia
Arousal at imagining oneself femaleA proposed explanation for some male-to-female transgender identification, particularly among people whose gender dysphoria begins later and who are attracted to women. The theory describes sexual arousal associated with imagining oneself as a woman. The book notes that the concept and its explanatory scope remain contested.
Regulatory capture
Industry influence over its regulatorA situation in which a government agency tasked with regulating an industry becomes unduly influenced by the firms or professionals it oversees. Huemer argues that close ties between regulators and industry can lead rules to protect established providers or limit competition rather than primarily serving consumers.
Ideological defense system
Beliefs that shield ideology from challengeHuemer’s term for arguments or assumptions that make a political belief system resistant to evidence. Examples he discusses include treating questions as immoral, labeling dissent harmful, or dismissing critics based on identity. Such defenses can prevent open evaluation regardless of whether the protected beliefs are true.
Gender
Psychological traits associated with sexHuemer rejects defining gender solely as a social role or equating it completely with biological sex. He proposes that masculine and feminine gender consist of psychological traits associated with male and female biology. This is the author’s theoretical account, not an uncontested definition.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.