Calling Bullshit
The Art of Skepticism in a Data-Driven World
Carl T. Bergstrom and Jevin D. West · 2020 · Science
11 min · 11 chaptersNarrated in your voice
What it's about
Bergstrom and West argue that the modern flood of numbers, charts, and statistics has made bullshit easier to produce and harder to detect than ever before. The book teaches readers to recognize misleading claims built from real data — not outright lies, but numbers arranged to deceive or impress. It matters because quantitative bullshit now shapes public health decisions, elections, and business strategy, and traditional media literacy training rarely covers how to interrogate a graph, a study, or an algorithm.
Reading plan
Pick a finish date and Genius lays out the days — the plan shows today's target and keeps you honest.
Read together
Start a circle and share the code — everyone sees everyone's honest place in the book. Accountability, not leaderboards.
Who it's for
- A professional who reads reports and dashboards and wants to spot when the numbers are being spun - A student or citizen trying to evaluate news claims and social media statistics without a statistics degree - A manager or journalist who needs practical tools to question data before repeating it
What’s inside
- 1What Bullshit Is1 min
Bullshit is distinct from lying because it aims to impress or persuade through the appearance of rigor rather than through factual falsehood, which makes it harder to refute than a simple lie.
- 2Causation, Correlation, and the Stories We Want to Believe1 min
Most causal claims in public discourse are unsupported correlations dressed up as causation because a causal story is more satisfying and more shareable than an honest 'we don't know.'
- 3How Numbers Lie Without Lying1 min
Selectively chosen statistics — like using the mean instead of the median, or cherry-picking a favorable time window — can make a technically true number tell a false story.
- 4Visualizations That Deceive1 min
Charts and graphs carry unearned authority as objective evidence, so small design choices — truncated axes, misleading scales, or cherry-picked comparisons — can distort perception even when every plotted number is accurate.
- 5Selection Effects and the Data You Never See1 min
Conclusions drawn only from surviving, visible, or reported cases are systematically biased because the missing data — who didn't respond, what didn't survive, what wasn't published — is never random.
- 6Algorithms and the New Machinery of Bullshit1 min
Machine learning systems can generate bullshit at industrial scale because they optimize for patterns in training data without any mechanism for caring whether their outputs are true, fair, or meaningful.
- 7Refuting Bullshit in Public1 min
Debunking a false claim often backfires or fails to spread because true corrections are boring and lack the narrative punch that made the original bullshit go viral in the first place.
- 8Building the Habit of Skepticism1 min
Calling bullshit effectively is a practiced skill requiring specific habitual questions, not innate intelligence, which means anyone can improve at it through deliberate practice regardless of technical background.
About Carl T. Bergstrom and Jevin D. West
Carl T. Bergstrom is a biology professor at the University of Washington studying information and evolution. Jevin D. West is an information scientist at the same university, co-directing the Center for an Informed Public. Both created a popular university course on spotting bullshit, which became the basis for this book, drawing on backgrounds in quantitative science and network analysis.
