CaddieOpen in CaddieDownload for Mac

How to argue without preaching

Shared by Caddie Library · Updated
Chapters show their main points
Chapter 17 of 20 · Structure and Argument
← The growth course
1. Finding Product-Market Fit
Test for real customer pull earlyPick a narrow niche to start inLaunch a simple first version fastLearn from users and customersRead early signs of product-market fit
2. Onboarding and Activation
Structure onboarding as a storyAsk and use onboarding questionsGet users to the aha moment fastDecide how long onboarding should beBuild quiz funnels for paid traffic
3. Paywalls and Pricing
When to choose a hard paywall over freemiumDesign a paywall that convertsPrice annual and monthly plansSet, raise and localize pricesRun free trials and free tastes
4. Business Models and Revenue
Structure subscription tiersKeep a free tier that pays offCharge for usage and AI creditsPair hardware with a subscriptionAdd ads, affiliates and partners
5. Retention and Engagement
Measure retention honestlyBuild habit loops that bring users backUse notifications and lifecycle messagesReduce churn and handle cancellationsBuild social loops and community
6. Brand and Positioning
Position against bigger incumbentsName and describe your productEarn trust in high-stakes categoriesBorrow credibility from celebrities and experts
7. Organic Content and Creators
Find, vet and recruit creatorsPay creators fairly and safelyBrief and manage creatorsFind and copy viral formatsScale one format across many accountsRank in AI search and SEO
8. Paid, App Store and Viral Growth
Make ads that convertTurn organic winners into paid adsStructure and scale Meta campaignsTell if paid acquisition is profitableWin in the App StoreBuild viral and referral loops
9. Experimentation and Analytics
Run trustworthy experimentsJudge test results correctlyRun more experiments fasterWhat metrics to track and how to define them
10. Building AI Products
Build for the models of tomorrowDesign AI interfaces beyond chatDesign AI agents people can trustMeasure AI quality with evalsControl AI costs and pricingBuild products with AI coding tools
11. Teams, Hiring and Funding
Hire people who thrive on ambiguityOrganize small product teamsRun exploratory bets and kill themWork as a team with AI agentsWhen to raise money or sell your company
12. Founder Principles
Pick a mission worth building a company aroundWhy to trust your own judgment over the consensusTurn failure and setbacks into the next stepLearn from rivals and copy what works
13. Choosing Where to Compete
Make low cost your competitive advantageCreate a market instead of fighting for shareWhy scarcity and brand can beat volumeBuild an advantage that feeds itselfAttack a bigger rival where it cannot followWhen to stay focused on the core business
14. Building and Leading the Company
Hire great people and give them roomAlign pay and ownership with the missionKeep your culture and standards as you growBuild candor and fast decision-makingWhy leaders must stay close to the work and customersChoose and prepare a successor
15. Capital and Deals
Fund growth without losing controlAcquire companies and keep what makes them workWhat to do with cash: reinvest, buy back or holdWhy to keep a cash cushion and avoid heavy debtStructure ownership to protect your missionNegotiate deals that keep your leverage
16. Finding Ideas and Thinking
How to find ideas worth writing aboutWhy writing is how you find out what you thinkHow to research so the writing carries feelingHow to read like a writer
17. Structure and Argument
How to open so readers keep goingHow to find the shape before you draftHow to keep readers curious with questions and cluesHow to argue without preachingHow to end so the last page lingers
18. Clear Sentences and Voice
How to write plainly for a reader who knows nothingHow to choose specific details over generic onesHow to sound like yourself on the pageHow to shape rhythm and phrases people remember
19. Characters and Storytelling
How to build characters readers care aboutWhat makes a scene move: want, obstacle and changeHow to write dialogue that sounds real but stays shortHow to make readers feel without telling themHow to turn information into a story
20. Revising, Habits and Readers
How to cut and rewrite your own draftHow to get feedback you can useHow to build a writing routine that lastsHow to get unstuck when a draft stallsHow to use AI without losing your voiceHow to write for one specific reader

1 The bottom line

Make the case through story, fair opposing views and checkable evidence, then step back and let readers reach the conclusion. A stated lesson ends their thinking, while a case they assemble themselves is one they believe.

2 Key takeaways

Institutions pick 15% explained over 20% unexplained
Clear explanation15%
No explanation20%
Per Will Storr, a story about why returns will continue outweighs five extra points of return.
  • Let a story, scene or image carry the argument. People think in stories and believe in them.
    • Will Storr1:01:40 says institutions prefer 15% returns with a clear explanation over 20% returns nobody can explain.
    • Robert Greene1:13:20 walks readers through ancient Babylon so they ask why our cities are not like that.
  • Give the other side its strongest form before you disagree. Readers trust a writer who has faced the best objection.
    • Chris Dixon28:51 rewrote his objections section to grant each one its strongest version and draw design lessons from it.
    • Derek Thompson13:24 states even ideas he finds horrific in a form their holders would recognize.
  • Show your bias and mark fact apart from interpretation. Hiding your position withholds data the reader needs.
  • Do not state the lesson. A stated point ends the reader's thinking.
    • Paul Harding50:28 says that if readers get the point, they never have to think about it again.
    • Robert Greene26:05 shows what "crush your enemy totally" meant in history without calling it evil, because condemning it would treat readers like children.
  • Attack your own argument before readers do. The easiest person to fool is you.
    • Daniel Pink44:40 says taking both sides, as in moot court, also trains you to anticipate the other side's argument.

you don't tell the reader what to think. You present them with the good and the bad and the ugly and let them decide.

Nick Bilton, author of American Kingpin

2.1 Stress-test the argument

Argue against yourself first, so the draft already survives the other side.

  • Practice taking both sides. Daniel Pink44:40
    • Pink says the most useful thing he got from law school was arguing either side, as in moot court where a coin flip assigns yours.
    • David Perell adds Tyler Cowen's exercise: write the most compelling argument you can for something you vehemently disagree with.
    • Ezra Klein54:15 suggests writing a case for something and a case against it. It feels weird to the audience and is hard, but it keeps you from talking yourself out of doubt too early.
  • Treat your first draft as a claim to test. John Lennox29:32
    • Lennox follows Feynman's rule of bending over backwards to criticize your own work. He pairs it with the misunderstanding question and with outside readers.
    • Ryan Holiday21:50 found that rereading old books made him cringe at certainty, not style. For new editions he cut blunt statements and added what he wished he had known.
  • Pick the question people are asking and commit to an answer. Dan Wang40:35
    • Wang identified the questions people were asking about China that year and offered a real answer, knowing he might be wrong.
    • Scott Galloway53:48 tells his team the goal is to catalyze a conversation. He backs a controversial topic with data and makes the argument broad enough that people who disagree still take something from it.
    • Wang's 2021 letter took on three questions: the tech crackdown, how zero COVID would end, and what common prosperity meant.
    • Galloway looks for a sacred cow and argues against his own group's interest. He told wealthy executives he paid the same 9,000 dollars in social security tax as someone earning 160,000, despite earning 16 million Scott Galloway51:52.
  • Set the test the finished piece must pass. Ted Gioia11:26
    • Gioia asks whether the piece is fair, accurate and persuasive to fair-minded readers. If not, he counts it a failure.
    • Howard Marks1:12 aims for the reaction "I never thought of it that way". A piece that only matches what readers already believe contributes little.
    • Gioia estimates maybe 50% of newsletter and article writing now carries an agenda, which makes readers rightly skeptical.

2.2 Open on shared ground

Win agreement on the problem before arguing anything
The reader says 'yeah' to the problem first, so the pitch lands on someone already listening.

Get the reader to agree with you on the problem or the goal before you argue anything.

  • Open with vulnerability, then a shared goal, then evidence. Johann Hari50:19
    • Hari learned from deep canvassing at the LA LGBT Center that you connect emotionally before you argue.
    • He then names a goal the reader already shares, such as less addiction or less depression.
    • Sperber and Mercier's work finds that juries weigh facts well because they agree on their goal. Lost Connections opens with Hari nearly dying from a poisoned apple in Vietnam.
  • Get the reader nodding at the problem before you pitch. Jason Fried47:02
    • Fried first describes how things used to be and what went wrong, until the reader agrees. Only then does he introduce the new reality.
    • Readers who have said "yeah" are open to what comes next.
    • The Once and Hey letters share one structure: how it was, what broke, what we will do about it. Great product demos use the same order.
  • Let readers commit first, then turn it on them. Mark Normand29:14
    • Normand says Louis CK is the king of the moment where you laugh and he then points out you laughed at something that did not bother you, though a similar thing does.
    • Setting up the premise early gets the reader involved, and then the reveal lands.
    • Normand has listened to that bit 20 or 30 times and calls it extremely well constructed.

2.3 Let a story, scene or image do the arguing

Make the point through something the reader can see, so the contrast does the work.

  • Earn every insight with a story, and keep the story complicated. Johann Hari51:45
    • Hari opens Chasing the Scream with Billie Holiday and Harry Anslinger instead of stating that the drug war began for racist reasons.
    • He refuses to flatten people. A two-dimensional, morally unambiguous story does not earn the insight.
    • Will Storr1:01:40 says to give your pitch a story about what is good, what we want and how we get there.
    • Hari does not present Holiday as a magical saint, because that would not be true.
    • An investor reports that institutional money favors the fund with the clearer story of why its returns will continue Will Storr1:01:40.
  • Answer an accusation with a story, never a statistic. Lulu Cheng Meservey58:22
    • Accusations arrive as stories and defenses arrive as statistics, so the defender loses.
    • Pair your data with a human case. Whoever writes the first draft of the frame sets the criteria everyone else reacts to David Perell23:07.
    • In the free trade debate, detractors tell of a laid-off corn farmer while supporters cite productivity up 4%.
  • Walk the reader through a scene, or set two passages side by side. Robert Greene1:13:20
    • Greene takes the reader down the processional path to the Ishtar Gate, past the ziggurat and the Hanging Gardens. The reader alone asks why our world is not like that.
    • Tim Mackie and Jon Collins21:26 point to biblical authors who give two stories and leave the meaning to the contrast. Adam and Eve see, desire and take the fruit, and David sees, desires and takes a woman from his roof.
    • If your point is abstract, show it in a person. Riva Tez28:33 says qualities seen in a character's actions become richer and more alive than a dictionary word.
    • Greene did a lot of research for the history chapter to build the walk.
    • Tez says people who annoy her now remind her of Lillian Rearden from Atlas Shrugged, which gave her a framework for understanding them Riva Tez28:33.
  • Use a metaphor every reader has lived, and check that it is fair. Michael Pollan39:30
    • Pollan set a sunlight-based food chain against an oil-based one and let readers choose.
    • Jason Fried42:42 framed subscription software as renting, with the vendor as landlord. He ended the paragraph with "Boom, you're evicted."
    • Pollan warns that metaphors imply an equation, so check yours is accurate, not only vivid. The brain-as-computer metaphor misled people into assuming conscious AI is easy.
    • For heavy subjects, Yann Martel18:20 says animal characters give distance, as in Animal Farm. A talking animal announces that the story is a metaphor.

2.4 Give the other side a fair voice

State opposing views so their holders would recognize them, then respond to the real argument.

  • Steelman before you disagree. Derek Thompson13:24
    • Thompson constructs the steel man of an argument, then explains why he thinks it is wrong. That gives him tension without cheap attacks.
    • In his first draft, Chris Dixon28:51 raised objections and simply said they were wrong. In the rewrite he granted the strongest version of each.
    • Security researcher Moxie Marlinspike's criticism became "that's a good argument, here is what to take from it". Dixon also softened overclaims, saying blockchains allow a richer design space for governance, not better governance Chris Dixon28:51.
  • Let the other side speak in its own voice. Jodi Picoult41:32
    • Picoult says most people form opinions young and never ask why. She asks readers to listen, then ask why they believe what they believe.
    • Small Great Things has a white supremacist narrator. For Sing You Home she interviewed Focus on the Family even as her son had just come out.
  • Show that you have read the opposing arguments. Marc Andreessen1:06:51
    • Andreessen wants informed readers to think they may or may not agree, but he understands the topic. He responds to the actual arguments rather than skipping them.
    • Instead of citing everything, he plans to end his 30 to 40 page manifesto with a reading list that documents his claims.
  • Make readers of other views feel welcome. Tyler Cowen and Alex Tabarrok31:32
    • Cowen says many columnists lay down the line in a take-it-or-leave-it way. He wants readers of different politics, religions and countries to feel their perspective is welcome.
    • Living in different countries helped him see his own biases, which you cannot remove but can encircle.

2.5 Keep the evidence honest

Tell readers where you stand, what is fact, and how sure you are, so they can weigh your case.

  • Tell readers your biases in the introduction. Diarmaid MacCulloch36:36
    • Readers can then weigh a passage by thinking "he would say that, wouldn't he?", which makes them more secure.
    • MacCulloch says that without this you deprive readers of essential data. He does it in the introductions to his books, including the one on Christianity.
  • Separate fact from interpretation. Chamath Palihapitiya18:57
    • Mark where objective material ends and your opinion begins, and label the subjective sections. Readers can discard your opinion and keep the facts.
    • When the blend is unclear, the reader has to work out whether they are being informed or steered Chamath Palihapitiya20:44.
    • He cites a study finding only about 30% of papers in Cell, Science and Nature were reproducible, as proof that fact and opinion get confused even in trusted places.
  • State your degrees of confidence, and add the "but". Diarmaid MacCulloch17:50
    • Where sources survive only as fragments, MacCulloch says so, with phrases like "we may think it is quite likely".
    • When you see a simple, heroic version of events, add: yes, I see that, but I also see this Diarmaid MacCulloch15:55.
    • A novelist can fill gaps silently. Hilary Mantel invented a romance in Antwerp for Cromwell to account for a known illegitimate child.
  • Write claims that could be proved wrong, with logic that holds on a reread. Harry Dry5:58
    • A claim that can be wrong puts your head on the chopping block, and readers sit up.
    • Howard Marks37:04 says you cannot bluff in writing, because the reader will reread and notice when something does not follow.
    • "Worn by supermodels in London and dads in Ohio" is checkable. "Change an entire industry" is not.

2.6 Step back and cut the lecture

Once the case is on the page, remove every line that tells the reader what to conclude.

  • Skip the takeaway and describe without explaining. Paul Harding50:28
    • Harding says a message like "be kind to strangers" is received opinion. He brings the reader to a moment and says look what I found.
    • A novel works by recognition, not argumentation Paul Harding53:40.
    • He knew only that a character in Tinkers left his family, so he wrote what it was like for the one who left and the people left behind. "To describe it doesn't mean I'm explaining it" Paul Harding1:07:21.
  • Present the good, the bad and the ugly, and let readers judge. Nick Bilton1:13:23
    • Bilton deliberately gave no verdict on Ross Ulbricht in American Kingpin.
    • Robert Greene26:05 wants readers to reach a conclusion on their own, whether the same as his or the opposite.
    • Bilton is asked constantly whether Ulbricht deserved his sentence, and everyone gives a different answer.
    • Greene says you cannot change people's morals with a book, only give them information and stories Robert Greene26:05.
  • Let the theme come in sideways. Andrew Hunter Murray14:00
    • Murray calls it sugarcoated broccoli. The theme sits below the character and the plot.
    • Scott Dikkers48:43 says to pick a small subtext and make the joke the main thing. "As soon as you're tasting the medicine, you've lost."
    • Elif Shafak35:59 opens a space of nuance and plurality, then steps back so readers create the meaning with her.
    • Murray says Breaking and Entering could have been a dry book about housing shortages in Greater London, which no one would want.
    • Dikkers praises Swift's A Modest Proposal but says the end of Gulliver's Travels is terrible because Swift preaches Scott Dikkers48:43.
  • Cut lines that lecture, and do not point the finger at readers. Jack and Ryan of AJR15:15
    • Ryan wrote "that's the way it always is and that's the way it's always been." Jack felt lectured at. The fix was a short, felt line: "oh my god, that's so insane."
    • Tucker Max1:10:20 says memoirs often fail when the author tells their story while also telling readers what to do.
    • Rob Henderson1:21:46 relays Max's advice to write from your scars, not your wounds. Wounded writers still want to blame someone, and it reads like grudge-settling.
    • The simpler line stayed in the song that became AJR's biggest hit, with over 170 million video views.
    • Henderson says he can usually tell from the first chapter how honest a memoir is being Rob Henderson1:21:46.

2.7 What the books say

The book adds a rule for fair opposing views: bring up alternatives only when they help the reader, never as targets to knock down.

  • Mention an alternative only if the reader already knows it. Barbara Minto, The Minto Pyramid Principle
    • Place that alternative in the Complication, where it is part of the problem the reader already faces.
  • Recommend your choice because it solves the problem, not because the others fail. Barbara Minto, The Minto Pyramid Principle
    • If a reader insists on seeing the options, put them in an appendix chart that compares them against criteria.

2.8 Where writers disagree

Writers split on whether to deliver a verdict
No verdictMiddle pathCommit
WhoNick BiltonScott GallowayDan Wang
ApproachShow good, bad and uglyStart a conversation, not be rightGive an answer, not just map issues
Reader's roleJudges aloneTakes something even if disagreeingWeighs a stated position
Ezra Klein adds that few writers manage to convey uncertainty without falsely resolving it.

Writers split on whether to give a verdict, where lessons belong, and whether a message may cost you the laugh.

  • Verdict or no verdict. Nick Bilton1:13:23 gave none on Ross Ulbricht.
    • Dan Wang40:35 urges writers to commit to an answer instead of only mapping the issues.
    • Ezra Klein54:15 wishes he were better at writing uncertainty without falsely resolving it, and says few writers do it well.
    • Galloway's middle path: aim to start a conversation, not to be right, and keep the argument broad enough that disagreers still take something Scott Galloway53:48.
  • Lessons on the page or off it. Paul Harding50:28 would call stating a point fraudulent.
    • Tucker Max1:10:20 says that if you want lessons, put them in a separate section at the end.
    • Alain de Botton1:16:56 says a lecture gives information, but a sermon adds story and feeling so behavior changes. He places himself on the sermon side.
    • David Goggins wanted lessons in his book, so they went after each chapter in a separate section. Max says that made the audiobook work so well Tucker Max1:10:20.
    • Greene calls himself a big believer in sneaking in messages, which he does through the scene, not the statement Robert Greene1:13:20.
  • Story or chart. Lulu Cheng Meservey58:22 says fighting a story with a statistic means you are always losing.
    • David Perell1:19:05 says a chart can make your argument "in the snap of a finger". He likes that Claude lets you describe the chart in natural language instead of building it in Excel.
  • Message versus the joke. David Zucker43:45 defines axe grinding as making a point at the expense of the film. You violate that rule at your peril unless you are funny enough to overcome it.
    • His own American Carol, which mocks the left, breaks the rule on purpose. He skips topical political jokes, such as 1970s Nixon sketches and SNL Trump jokes, as a mix of axe grinding and straw dummy.

Sources

Part of Structure and Argument in the growth library.