The Paris Review notes: “In Zach Williams’s “Trial Run” (issue no. 239, Spring 2022), an employee is subjected to two coworkers’ conspiracy theories when their office is targeted by an anonymous white supremacist hacker.”
I loved everything about this story except the ending. ChatGPT says:
In Trial Run, the protagonist, an office worker, finds himself snowed in at his Manhattan office building, accompanied only by Manny, the conspiracy-theory-obsessed security guard. As the storm intensifies, Manny’s increasingly erratic behavior and cryptic remarks about global conspiracies unsettle the protagonist. The story concludes with the protagonist’s escape from the building, leaving behind a lingering sense of unease about what is real and what might be paranoia fueled by isolation.
Themes and Interpretations
Isolation and Paranoia: The snowstorm and the empty office setting amplify the protagonist’s sense of isolation, making him more susceptible to doubt and paranoia. Manny’s conspiracy theories blur the line between reality and delusion, reflecting the protagonist’s inner turmoil.
The Bookish Elf
The Facade of Normalcy: Manny’s sudden shift from a seemingly benign figure to a source of anxiety challenges the protagonist’s perception of normalcy. This transformation underscores the fragility of the social facades we maintain.
Unreliable Narration: The story’s ambiguity leaves readers questioning the reliability of the protagonist’s perspective. Is Manny truly a threat, or is he a projection of the protagonist’s fears? This uncertainty invites readers to reflect on the nature of truth and perception.
The Washington Post
Conclusion
The ending of Trial Run is deliberately unsettling, leaving readers with more questions than answers. This open-ended conclusion serves as a commentary on the complexities of modern life, where the boundaries between reality and perception are often blurred. Williams’ story challenges readers to confront their own assumptions and the tenuous nature of the world around them.
>>>It was a lousy payoff.
I completely understand your reaction! It’s really common to feel let down by an ending, especially when the rest of the story is so engaging and promising. In Trial Run, Williams builds up such a strong atmosphere of tension, isolation, and psychological unease that you’re probably invested in seeing how these elements resolve. So when the ending doesn’t offer a clear resolution or answers to the questions the story raises, it can feel like a missed opportunity.
The buildup with the conspiracy theories, the protagonist’s growing unease, and the strange dynamic between the two characters all create a strong sense of anticipation. If you were hoping for a more satisfying payoff—whether that’s a clear explanation of Manny’s behavior or a cathartic conclusion for the protagonist—the ambiguous ending can feel a bit unsatisfying.
It almost feels like a story where the resolution might be more about the emotional experience or the exploration of the protagonist’s psychological state than giving you a definitive answer. That’s something some readers really connect with, but if you’re someone who prefers closure and answers, it can definitely leave a bad taste.
Nathan and Dave Green debate this question. Grok says:
The North American Free Trade Agreement (NAFTA), implemented in 1994, had a mixed impact on the standard of living for the average American, with evidence suggesting modest overall benefits but significant variation across groups and regions. Here’s a concise breakdown based on available data and analysis:
Positive Impacts:
Economic Growth and Consumer Benefits: NAFTA increased trade between the U.S., Canada, and Mexico, boosting U.S. GDP by an estimated 0.5% to 1% over time (e.g., Congressional Budget Office, 2003). This growth came from lower trade barriers, increasing access to cheaper goods, particularly from Mexico. For consumers, this meant lower prices for items like produce, electronics, and clothing, effectively raising purchasing power.
Job Creation in Some Sectors: Export-oriented industries, such as agriculture, automotive, and services, saw job growth. U.S. exports to Mexico and Canada grew from $337 billion in 1993 to over $1 trillion by 2017 (USTR data), supporting jobs in these sectors.
Wage Effects in Certain Industries: Workers in export-driven sectors often saw wage gains due to increased demand. Studies (e.g., Hufbauer and Schott, 2005) estimate a net positive effect on U.S. wages, though small, averaging 0.1-0.3% higher by the 2000s.
Negative Impacts:
Job Losses in Manufacturing: NAFTA contributed to the displacement of roughly 700,000 U.S. jobs by 2010, primarily in manufacturing (Economic Policy Institute, 2011). Industries like textiles and electronics faced competition from lower-wage Mexican labor, hitting regions like the Rust Belt hard. Displaced workers often moved to lower-paying service jobs, reducing their standard of living.
Wage Suppression in Some Sectors: Increased competition from Mexican labor put downward pressure on wages for low-skill manufacturing workers. Autor et al. (2016) found that trade shocks from NAFTA and other agreements reduced wages for some blue-collar workers by 1-2% in affected regions.
Regional Disparities: Communities dependent on manufacturing (e.g., Michigan, Ohio) faced economic decline, with ripple effects on local services and infrastructure. Meanwhile, border states like Texas benefited from trade logistics and cross-border commerce.
Net Effect on Standard of Living:
Modest Gains for the Average American: Most studies conclude NAFTA had a small positive effect on the average American’s standard of living, driven by lower consumer prices and slight GDP growth. The U.S. International Trade Commission (2016) estimated a net welfare gain of $56 billion annually by 2016.
Uneven Distribution: Gains were uneven, with higher-skill workers, urban consumers, and export-sector employees benefiting more. Low-skill manufacturing workers and certain rural communities faced losses, with inadequate retraining programs exacerbating the impact.
Broader Context: NAFTA’s effects are hard to isolate from other factors like automation, China’s rise in global trade, and domestic policy changes, which also influenced jobs and wages.
Conclusion:
NAFTA likely improved the standard of living for the average American by a small margin through cheaper goods and economic growth, but it harmed specific groups, particularly low-skill manufacturing workers in certain regions. The net positive effect masks significant losers, highlighting why perceptions of NAFTA remain polarized.
Nathan says that changing the composition of the population is a moral question. Well, there’s a moral question too in what type of economy we want? Do we want an economy run for maximum economic efficiency according to certain economic models or do we start with other goals?
Some cite efficiency as a rationale for free trade. For adherents of this faith, the sole objective of trade policy is market efficiency. Lower tariffs and nontariff barriers reduce the costs of producing and distributing goods and services; that, in turn, makes society as a whole better off — or so the argument goes. Of course, there are advantages to lower prices when considering Americans as consumers. The problem is that adherents of this view seldom consider how this single – minded pursuit of efficiency affects the men and women who actually do the producing.
According to the definitions preferred by these efficiency – minded free traders, the downside of trade for American producers is not evidence against their approach but rather is an unfortunate but necessary side effect. That’s because free trade is always taken as a given, not as an approach to be questioned. Rather than envisioning the type of society desired and then, in light of that conception of the common good, fashioning a trade policy to fit that vision, economists tend to do the opposite: they start from the proposition that free trade should reign and then argue that society should adapt. Most acknowledge that lowering trade barriers causes economic disruption, but very few suggest that the rules of trade should be calibrated to help society better manage those effects. On the right, libertarians deny that these bad effects are a problem, because the benefits of cheap consumer goods for the masses supposedly outweigh the costs, and factory workers, in their view, can be retrained to write computer programs. On the left, progressives promote trade adjustment assistance and other wealth – transfer schemes as a means of smoothing globalization’s rough edges.
Nathan Cofnas writes: “There are currently 6.8 million unemployed Americans and 8 million job openings.”
Grok says:
Nathan Cofnas’s claim that there are “6.8 million unemployed Americans and 8 million job openings” suggests a labor market with more opportunities than unemployed workers, implying a relatively strong economy. This aligns with government statistics like the U-3 unemployment rate, which was reported at 4.2% in November 2024, indicating low unemployment. However, Eugene Ludwig’s Politico article argues that such statistics misrepresent economic reality, painting an overly optimistic picture that contrasts with public perception and lived experiences.
Comparison
Unemployment Metrics:
Cofnas: Cites 6.8 million unemployed, consistent with official U-3 data, which counts only those actively seeking work and not discouraged workers or the underemployed. The 8 million job openings suggest a tight labor market where jobs outnumber job seekers.
Ludwig: Critiques U-3 as misleading because it counts underemployed workers (e.g., part-time workers seeking full-time jobs) and those earning poverty wages as “employed.” His research at the Ludwig Institute estimates a “functional unemployment” rate of 23.7%, including those who are underemployed, discouraged, or earning insufficient wages (e.g., below $25,000 annually). This starkly contrasts with Cofnas’s implied optimism, highlighting a much bleaker labor market for many Americans.
Economic Strength:
Cofnas: The surplus of job openings (8 million vs. 6.8 million unemployed) suggests economic vigor, as businesses are hiring and opportunities exist. This aligns with narratives of a robust post-pandemic recovery often cited by policymakers.
Ludwig: Argues that government metrics like GDP obscure disparities. While GDP growth indicates prosperity, it doesn’t reflect how gains are distributed. Ludwig notes that since 2013, only those with advanced degrees or in affluent regions (e.g., San Francisco, Boston) have seen significant gains, while lower-income and less-educated workers, or those in struggling regions (e.g., Youngstown, Ohio), have stagnated or declined. This challenges the idea that job openings alone signal broad prosperity.
Contrast
Perception vs. Reality:
Cofnas: His figures reflect government data, which Ludwig argues Democrats overly trusted, leading to a disconnect with voters who felt economically strained. Cofnas’s statement doesn’t address whether job openings are accessible, well-paying, or geographically relevant to the unemployed.
Ludwig: Emphasizes that public perception of economic decline is more accurate than government stats. He cites visible signs like homelessness and derelict regions, which resonate with voters’ experiences. His alternative metrics (e.g., functional unemployment, true cost-of-living increases) align with Republican skepticism of rosy official numbers, explaining voter frustration in 2024.
Wage and Inflation Insights:
Cofnas: Doesn’t address wages or inflation, focusing solely on unemployment and job openings. His data implies workers can find jobs, but not whether those jobs pay enough to meet living costs.
Ludwig: Highlights flaws in wage and inflation metrics. Official weekly earnings ($61,900 median) exclude part-time and unemployed workers, inflating perceptions of income. Ludwig’s estimate ($52,300 median) shows workers earn 16% less than reported. On inflation, the Consumer Price Index (CPI) underestimates cost-of-living increases for low- and middle-income households, who spend more on necessities like groceries and rent. Ludwig’s alternative indicator shows a 35% higher cost-of-living rise since 2001, with a 9.4% increase in 2023 alone, outpacing wage growth and eroding purchasing power by 4.3% that year.
Methodological Critique:
Cofnas: Relies on unadjusted government statistics without questioning their validity, reflecting the Washington insider perspective Ludwig critiques.
Ludwig: Advocates for new metrics to capture economic realities, like including underemployment and poverty wages in unemployment rates or focusing inflation measures on necessities. He argues that current methodologies, unchanged for decades, mislead policymakers and fail to address disparities.
Synthesis
Cofnas’s statement reflects the optimistic narrative of government statistics, suggesting a labor market with ample opportunities (8 million openings for 6.8 million unemployed). Ludwig, however, reveals the limitations of these numbers, showing that unemployment is functionally higher (23.7% vs. 4.2%), wages are lower ($52,300 vs. $61,900), and inflation hits harder for most Americans than the CPI suggests. While Cofnas’s data might appeal to those citing headline figures, Ludwig’s analysis explains why voters felt economically worse off, aligning with visible struggles like homelessness and regional decline. Ludwig’s call for updated metrics underscores the need to bridge the gap between statistical optimism and ground-level reality, a disconnect that Cofnas’s figures don’t address.
Dave Green is unbearable. He goes on and on how about the standard of living for the median American is now lower than it was in the 1990s. Grok says:
Yes, the standard of living for the median American is generally higher now than in 1995, but the picture is nuanced. Key factors include:
Income and Purchasing Power: Real median household income has risen from about $56,000 in 1995 to around $81,000 in 2023 (adjusted for inflation, 2023 dollars). Real per capita GDP has also grown significantly, from ~$39,000 to ~$65,000 (2023 dollars). This reflects higher purchasing power overall.
Technology and Goods: Access to advanced technology (smartphones, internet, streaming) and cheaper goods (electronics, clothing) has improved quality of life. Many products unavailable or costly in 1995 are now affordable and ubiquitous.
Healthcare and Life Expectancy: Life expectancy increased from 75.8 years in 1995 to 77.5 years by 2023, with better medical treatments and access to care, though disparities persist. Healthcare costs, however, have outpaced inflation.
Housing and Debt: Housing costs have risen faster than incomes (median home price: $147,000 in 1995 vs. $412,000 in 2023, adjusted). Student and consumer debt have also grown, straining some households.
Work and Leisure: Real wages for the median worker have grown modestly (~10-15% since 1995), but job polarization has reduced middle-skill opportunities. Remote work and flexible arrangements, enabled by tech, are more common, enhancing work-life balance for some.
Caveats: Inequality has widened, with top earners gaining disproportionately. Millennials and younger generations face challenges like delayed homeownership and higher debt burdens. Subjective well-being metrics (e.g., happiness surveys) show mixed results, with some reporting less optimism than in the 1990s.
Overall, material and technological improvements tilt toward a higher standard of living, but housing costs, debt, and inequality create trade-offs for the median American.
Dave Green says that GDP per capita is a poor proxy “longitudinally” for wealth. What does that mean?
Grok says:
When Dave Green says GDP per capita is a poor proxy “longitudinally” for wealth, he means it’s not a reliable measure for tracking changes in wealth over time. “Longitudinally” refers to analyzing data across multiple time periods to observe trends or changes, as opposed to a single snapshot (“cross-sectionally”).
GDP per capita, which is the total economic output of a country divided by its population, reflects average income or production at a given moment. However, Green’s point is that it doesn’t effectively capture how wealth—accumulated assets, savings, or net worth—evolves over time. Here’s why this might be the case:
Wealth vs. Income: GDP per capita measures income generated in a year, not accumulated wealth (e.g., property, investments, or savings). Wealth can grow or shrink independently of annual income, especially over long periods.
Inequality: GDP per capita is an average and doesn’t account for wealth distribution. Over time, wealth may concentrate among a few, while the majority’s wealth stagnates, even if GDP per capita rises.
Inflation and Cost of Living: Longitudinal comparisons of GDP per capita may not adjust well for inflation or changes in purchasing power, which affect real wealth.
Non-Market Factors: Wealth includes assets like real estate or inheritances, which GDP per capita doesn’t capture. These can significantly influence wealth trends over time.
Economic Shocks or Policies: Tax policies, recessions, or booms can disproportionately affect wealth accumulation (e.g., stock market gains or housing crashes) in ways GDP per capita doesn’t reflect.
In short, Green is arguing that GDP per capita misses key dynamics of wealth accumulation when you track it over time, making it a flawed metric for understanding long-term changes in economic well-being. For a more accurate picture, you’d need to look at metrics like net worth, wealth distribution, or asset ownership trends.
When we are told that there are 7.5 million job openings in America, how reliable is that figure?
Grok says:
The figure of 7.5 million job openings in America, typically sourced from the U.S. Bureau of Labor Statistics (BLS) Job Openings and Labor Turnover Survey (JOLTS), has some reliability but comes with caveats. Here’s a breakdown of its reliability based on available information:
Source and Methodology: The JOLTS data is collected from a sample of about 21,000 businesses and government establishments, representing a subset of the U.S. economy. The BLS uses statistical modeling to estimate total job openings, defined as positions open on the last business day of the month that meet specific criteria (work available, start within 30 days, active recruitment). This sampling approach introduces potential errors, as it relies on extrapolating from a limited dataset.
Response Rate Concerns: Posts on X highlight a significant issue: the JOLTS survey response rate is low, around 33% in recent years, meaning roughly 67% of the data is estimated or imputed. Low response rates can reduce accuracy, as the BLS must rely on assumptions or historical data to fill gaps, potentially skewing results.
Revisions: The JOLTS data is frequently revised, often downward. For example, February 2025 job openings were revised down by 88,000 to 7.5 million, and December 2024 data was revised down by 92,000 to 7.5 million. These revisions suggest initial estimates may overestimate openings due to incomplete data or modeling errors. Over the last 21 months, 16 reports were revised downward, indicating systematic overestimation.
Economic Context and Interpretation: The 7.5 million figure for February 2025 aligns with a broader trend of declining job openings from a peak of 12.2 million in March 2022. However, it’s still high historically (pre-2021, openings rarely exceeded 8 million). This suggests resilience in labor demand but doesn’t capture whether these openings are for high-quality, well-paying jobs or low-wage, high-turnover positions. Additionally, external factors like policy uncertainty (e.g., tariffs) or labor force participation (down from pre-COVID levels) may distort employer demand or reporting.
Skepticism from Observers: Some X posts question the data’s validity, calling it “guesswork” or “made up” due to the low response rate and frequent revisions. While these claims may exaggerate, they reflect legitimate concerns about the precision of estimates when two-thirds of the data is imputed.
Counterpoints: Despite limitations, JOLTS remains a widely used indicator, corroborated by other labor market metrics like unemployment rates (4.2% in April 2025) and payroll growth (177,000 jobs added in April). These suggest a solid, though cooling, labor market, lending some credibility to the job openings figure. The BLS also employs rigorous statistical methods to adjust for non-responses and seasonal factors, though these adjustments aren’t foolproof.
Conclusion: The 7.5 million job openings figure is a useful but imperfect estimate. Its reliability is undermined by low survey response rates, frequent downward revisions, and reliance on statistical imputation for a significant portion of the data. While it broadly reflects labor market trends, it may overstate or misrepresent the true number of openings, especially in specific industries or regions.
Nathan says that Americans are already poorer due to Trump’s tariffs.
Posted inEconomics|Comments Off on Did NAFTA improve the standard of living for the average American?
The debate between Dave Smith and Douglas Murray on Joe Rogan’s podcast has reignited a conversation on the right about the value of experts and expertise.
For those who didn’t catch the debate, here’s a brief summary. Its purpose was to ventilate the arguments for and against the West’s policy in Ukraine and Israel’s war in Gaza, with Murray representing those “for” and Smith those “against”. Right off the bat, Murray asked Rogan, “Do you think you’ve had enough people on who are supportive of either war?” (the implication being that he hadn’t). This prompted a lengthy discussion about whether Rogan had platformed too many people with non-mainstream or fringe views, and whether it’s even appropriate for non-experts (including Smith) to pontificate on subjects like geopolitics on one of the world’s most popular podcasts.
As I understood him, Murray was not arguing that non-experts shouldn’t be allowed to share their views, but rather that gatekeepers like Rogan have a responsibility to refrain from platforming them—or to only platform them alongside experts who have the knowledge to debunk their views.
“So what’s the solution—to not talk about it?” Rogan asks. To which Murray replies in an exasperated tone, “No, it’s to have more experts around.” He’s then challenged by Smith, who asks in relation to lockdowns and other Covid measures that he opposed, “Should I have just shut up?” Murray says, “No, no, no,” and explains that what he really means is, “Let’s have a bit of hygiene on our own side, not lift every sewer gate”. Murray also argues that it’s “weird” for people like Smith to talk “nonstop” about subjects in which they are not experts.
After the episode was published online, Murray received a lot of criticism: almost every single comment on the YouTube video makes fun of him in some way, and the top one has 37K likes. Yet a number of thoughtful commentators came to his defence. Konstantin Kisin, Nathan Cofnas and Yuki Zeman each penned an article defending Murray’s position on experts and expertise—i.e., that “certain people shouldn’t comment on certain things and that others should not elevate their voices” (to quote Kisin).
Like Kisin, Cofnas and Zeman, I obviously prefer to see people with a good understanding of the world have their voices elevated in the public discussion. (I have no interest in hearing from liars and ignoramuses.) However, I think that establishing who actually has a good understanding—i.e., who should have their voice elevated—is less straightforward than Murray and his defenders believe. And I think that by their own logic, Murray’s voice is one that shouldn’t be elevated.
All three pro-Murray articles take for granted that he is qualified to opine on the subjects he talks about. Beginning from this assumption, they then consider whether the targets of his criticism are similarly qualified, and conclude that some or most them aren’t. For example, Cofnas characterises Murray as someone who “went to Oxford, writes books with footnotes, and hangs out with academics”. This is in contrast to Smith, who “gets basic facts wrong in ways that reveal deep ignorance and intellectual irresponsibility”.
Now, I’m not disputing that Smith has made embarrassing errors. However, I don’t think these errors disqualify him from talking about the West’s policy in Ukraine or Israel’s war in Gaza—topics of which he seems to have a decent grasp based on his debate with Murray. Nor do they imply that Joe Rogan is irresponsible for having platformed him. After all, Murray too has made mistakes in his work, which I would contend are no less egregious than Smith’s. And I wouldn’t say he is unqualified to opine on these subjects.
Recently, the leftist journalist Nathan Robinson (who co-wrote a book with Noam Chomsky dealing with Israel/Palestine) penned a scathing review of Democracies and Death Cults, cataloguing numerous “errors, omissions and misleading statements”. Among the most egregious is that Murray used a fabricated quote and gave an anonymous Twitter account as the source. He also made false or misleading statements in a piece recounting his debate with Smith. For example, he stated that a previous guest on Rogan’s podcast had “very carefully tried to minimize the evil of Adolf Hitler” but that guest didn’t even mention Hitler.
More egregious than either Smith’s or Murray’s factual errors are some of the arguments Murray put forward on Rogan’s podcast. The first of these, which unsurprisingly went viral due to its patent absurdity, was Murray’s insistence that “if you’re going to spend a year and a half talking about a place, you should at least do the courtesy of visiting it”.
This is not a serious argument and I’m actually astonished that Murray and others have continued defending it.1 Despite claiming to have a “journalistic rule” of “trying never to talk about a country, even in passing, unless I’ve at least been there”, Murray frequently talks about countries he’s never been to. And so he should—because his supposed “journalistic rule” is bunk. There’s also the fact that Murray’s visits to Israel/Palestine are hardly impartial fact-finding missions where he makes a good-faith effort to get both sides of the story. It’s not even clear what his point is: many public figures have visited Israel/Palestine and come away with radicallydifferent views from Murray.2
The second of Murray’s bizarre arguments was his claim that “Paul Wolfowitz is a great figure for almost any deep conspiracy”, made in response to Smith’s point that General Wesley Clark had first heard about plans for US military intervention in countries like Iraq and Syria from Paul Wolfowitz.3 Murray goes on to state that “when you start talking about Paul Wolfowitz and Richard Perle” it is “awfully noxious smelling”, before quoting Mark Steyn’s quip that Wolfowitz’s name “starts with a nasty animal and ends Jewishly”.
Again, this is simply not a serious argument. In fact, it sounds like something a woke leftist would blurt out. To begin with, Smith is Jewish himself—and he brought up numerous non-Jewish figures during the discussion, including George Bush, Hillary Clinton and Barack Obama. What’s more, both Wolfowitz and Perle are described on their Wikipedia pages as “architects of the Iraq War”. To top it off, Smith barely even criticised the two men; he merely mentioned their names.
The third of Murray’s bizarre arguments was his accusation that Smith made an “appeal to authority” when he quoted the former head of the CIA. This is simply not what “appeal to authority” means. An appeal to authority is where you argue, “I am right because such-and-such an authority figure says so”. Smith merely cited a diplomatic cableas evidence to support a point he was making.4 If Murray were correct, it would be logically fallacious to ever cite diplomatic cables in arguments, ruling out entire scholarly disciplines.
The last of Murray’s bizarre arguments was his reference to Smith, made in a follow-up article for the New York Post, as someone “claiming some Jewish ancestry” (the implication being that he does not in fact have much Jewish ancestry.) This is just an ad hominem, Smith having confirmed that both his parents are Jewish and he is 84% Ashkenazi Jew.
To be clear, my point is not that Murray’s understanding of the world is so bad that his voice should not be elevated. It’s the opposite: despite the factual errors he’s made and bizarre arguments he’s employed, he is still qualified to opine on these subjects. And by the same standard, so is Smith. Of course, one could demand an even more exacting standard from podcast hosts, in which case both men might fall short. Though I personally think the quality of public discussion would suffer in that case.
Does this mean I support a total free-for-all where any lunatic can spout nonsense to millions of people? Not necessarily. I can imagine individuals whom I would regard it as inappropriate to invite on to a major podcast.5 However, I would generally favour a low bar for who counts as a “non-lunatic”. In other words, I would tolerate a wide gamut of speculation, conspiracy theorising and questioning of received wisdom.6
The first reason I would do so is that in fields like geopolitics, scope for genuine expertise is rather limited. This is not to say there aren’t people who could reasonably be called “experts”. There are. But the differences between them and non-experts are far smaller than in subjects like mathematics and the natural sciences.
Physicists can predict things to the tenth decimal place. Social scientists and political commentators can barely predict things better than chance. In a famous study, Philip Tetlock and his colleagues had 284 experts7 make precise, numerical forecasts8 about economic and geopolitical events over the span of two decades, and then scored them for accuracy. Remarkably, the experts performed no better than dart-throwing chimps and performed worse than simple algorithms (like extrapolating from the recent past). In addition, experts with specialist knowledge performed no better than those without such knowledge. For example, people who worked on the Soviet Union were no better at predicting things about the Soviet Union than people who worked on other topics.
Why are social scientists and political commentators so lousy at forecasting? It isn’t because they lack the smarts. (Many of the experts who took part in Tetlock and colleagues’ study were distinguished scholars who presumably had very high IQs.) It’s because human behaviour and human society are inherently less predictable—owing to the vicissitudes of human agency and the impossibility of running controlled experiments. Which is why there’s vastly more consensus in physics than there is in history or politics.
With an issue like the War in Ukraine, someone who’s reasonably intelligent and reads the news every day can have a meaningful conversation with an expert—especially about the ethical and strategic aspects. There are scarcely “right answers” to questions like, “Should the West be more or less hawkish toward Russia?” or “Should the West put more or less pressure on Ukraine?” And unsurprisingly, expert surveys reveal no consensuswhatsoever.
On a topic like electrodynamics, by contrast, there is literally nothing a non-expert can contribute to the discussion. And because there are experimentally-known “right answers” to all the relevant questions (except at the cutting edge of research), experts have reached an overwhelming consensus.
So, yes, it would be completely inappropriate for Rogan to invite someone like Smith on to talk about physics. But inviting him on to talk about geopolitics doesn’t seem inappropriate at all. We shouldn’t pretend that expertise in geopolitics is like expertise in physics.
The second reason I would tolerate a wide gamut of opinion is closely related to the first. In fields like geopolitics, not only are differences in genuine expertise much more compressed than in mathematics and the natural sciences, but it’s much harder to establish who has more and who has less.
Take John Mearsheimer. He is a professor of political science at the University of Chicago and has authored plenty of scholarly books and articles. At the same time, he is frequently denounced as a “Putin apologist” on social media—including by other members of his profession. The man even stands accused of distorting history to whitewash Putin’s crimes. Is he someone that it would be responsible to platform on a major podcast? Well, if you agree with his views, you’ll almost certainly say “yes”. But if you don’t, you might consider him a charlatan or worse.
There are numerous figures like this on all sides of every major issue. And there’s no surefire way to tell the real experts among them from the cranks, hacks and charlatans. This is not only because human behaviour and human society are inherently hard to predict, but also because subjects relating to political affairs call for moral judgements—and those moral judgements inevitably colour our interpretation of the facts.
You may have noticed that when it comes to the War in Ukraine and the Israel/Palestine conflict, there are very strong correlations between people’s views about factual matters (such as the causes of each conflict) and their moral judgements (such as what policy should be adopted). You almost never find someone who believes the War in Ukraine stems from Western meddling but who wants the West to supply Ukraine with long-range missiles. Or who believes the Israel/Palestine conflict is due to Palestinian intransigence but who wants the US to cut off aid to Israel. These correlations may arise in part because exposure to different facts leads people to different moral judgements. But they mostly arise because people’s pre-existing moral judgements lead them to different interpretations of the facts.
How are people with vastly different pre-existing moral judgements going to reach agreement about who’s a true expert and who’s a charlatan? Credentials might have worked in the past. But they aren’t going to cut it any more. After all, dazzling credentials can go hand-in-hand with stunning hubris and profound ignorance.
Who has a better understanding of the world? An educated layman who’s instinctively sceptical of transplanting Jeffersonian democracy to the Middle East, or the intellectual advocates of “nation-building” and “regime change”. Members of the latter group would certainly boast more impressive qualifications. They could surely reel off more facts about the relevant countries. And they’d undoubtedly do a better job at explaining concepts like checks on the executive. However, I think that any honest assessment of the evidence would show they have significantly worse understanding of the world. As cynics are fond of saying: the US spent $2 trillion and 20 years to replace the Taliban with the Taliban.
Which illustrates a point that Nassim Taleb has made about gauging expertise: we shouldn’t focus on the accuracy of people’s beliefs so much as the expectation—i.e., the accuracy weighted by the consequences of being right or wrong. As Taleb notes, if a trader predicts all the minor zigs and zags in the market but then loses millions when the market crashes, he will wind up poorer than one who fails to capitalise on the small fluctuations but profits handsomely during the crisis. By way of analogy, a commentator who makes many little errors but gets the big calls right may have a better understanding than one who give the appearance of erudition while advocating disastrous and costly policies.
I agree with Douglas Murray about almost everything, but he’s a polemicist, not a scholar. Grok notes:
Nathan Cofnas’s description of Douglas Murray—as an Oxford-educated writer who produces footnoted books and associates with academics—focuses on Murray’s formal credentials and social circle. It suggests a veneer of scholarly respectability. However, your view of Murray as a “predictable polemicist” without academic or scholarly credibility highlights a different angle: his work is often seen as provocative, prioritizing rhetorical flourish over rigorous scholarship.
Murray’s books, like The Strange Death of Europe or The War on the West, cite sources and engage with intellectual debates, which aligns with Cofnas’s portrayal. Yet critics, including some academics, argue his arguments lean on selective evidence and ideological agendas, lacking the nuance or methodological rigor expected in scholarly circles. His public persona—sharp-witted, combative, and media-savvy—further cements his reputation as a polemicist rather than a disinterested scholar.
Cofnas’s characterization isn’t entirely wrong but feels incomplete, emphasizing Murray’s polished credentials while sidestepping the polarizing, non-academic nature of his output. It’s like describing a fox as a well-groomed canine—it’s true, but misses the cunning.
Posted inExpertise|Comments Off on Noah Carl: In Defence of Non-Experts Establishing who should have their voice elevated is less straightforward than [Douglas] Murray and his defenders believe
"This guy knows all the gossip, the ins and outs, the lashon hara of the Orthodox world. He’s an [expert] in... all the inner workings of the Orthodox world." (Rabbi Aaron Rakeffet-Rothkoff)
