What Is a Logical Fallacy?
A logical fallacy is a flaw in reasoning. The argument sounds persuasive, but the logic doesn't hold.
Fallacies are dangerous because they feel convincing. Your brain doesn't naturally distinguish valid reasoning from invalid reasoning — it distinguishes "arguments I like" from "arguments I don't like."
Learning fallacies gives you a vocabulary for what's wrong with bad arguments. Instead of vaguely sensing something is off, you can name it precisely.
But here's the catch: it's easy to spot fallacies in arguments you disagree with. The real skill is spotting them in arguments you agree with.
The Major Fallacies
Ad Hominem (Attack the Person)
What it is: Attacking the person making the argument rather than the argument itself.
Example: "Why should we listen to his climate research? He drives an SUV."
Why it's wrong: The arguer's character, hypocrisy, or background doesn't determine whether their argument is valid. Bad people can make good arguments. Good people can make bad arguments.
The trap: Sometimes character does matter — like when evaluating credibility of testimony. The fallacy is when character is used to dismiss the argument itself rather than assess the person's reliability.
AI prompt to practice:
Here's an argument I encountered:
[Paste argument]
Is this attacking the argument or attacking the person? If it's ad hominem, how would I redirect to the actual argument?
Straw Man
What it is: Misrepresenting someone's position to make it easier to attack.
Example:
- Person A: "We should have reasonable regulations on gun ownership."
- Person B: "So you want to take away everyone's guns and leave people defenseless?"
Why it's wrong: You're not actually engaging with what the person said. You're fighting a caricature you created.
The trap: Sometimes we straw man unintentionally because we genuinely misunderstand the other position. This is why steelmanning (Chapter 6) is so valuable.
AI prompt:
I'm about to respond to this argument:
[Paste argument you're responding to]
My response: [Your response]
Am I representing their position fairly? Or have I created a straw man?
Appeal to Authority
What it is: Claiming something is true because an authority figure says so.
Example: "This celebrity endorses this diet, so it must work."
Why it's tricky: Authority does matter. Experts are more likely to be right than non-experts. But authority isn't proof. Experts can be wrong, can be outside their expertise, or can have conflicts of interest.
When authority is reasonable: When the authority is an expert in the relevant field, when experts generally agree, and when you've considered potential biases.
When it's fallacious: When the "authority" isn't expert in this area, when you're using authority to shut down rather than inform debate, or when you're cherry-picking the authorities who agree with you.
AI prompt:
Someone cited [authority/expert] to support [claim].
Help me evaluate:
1. Is this person actually an expert in this specific area?
2. What do other experts in this field say?
3. Are there potential conflicts of interest?
4. Does the actual evidence support the claim, regardless of who said it?
False Dilemma (Either/Or Fallacy)
What it is: Presenting only two options when more exist.
Example: "You're either with us or against us."
Why it's wrong: Reality usually has more than two options. False dilemmas force you to choose between extremes, ignoring the middle ground or alternative paths.
Why it works: Binary choices feel clear and decisive. Nuance feels weak. So false dilemmas are rhetorically powerful even when logically weak.
AI prompt:
I'm being presented with this choice:
[Either X or Y]
What other options might exist that aren't being presented? What's being left out?
Slippery Slope
What it is: Claiming that one thing will inevitably lead to extreme consequences without showing the causal chain.
Example: "If we allow this small regulation, soon the government will control everything."
Why it's tricky: Sometimes slopes are actually slippery — there are real causal chains where A leads to B leads to C. The fallacy is asserting the chain without evidence.
How to evaluate: Ask whether each step in the chain is actually likely. Are there stopping points? Are there mechanisms that would prevent escalation?
AI prompt:
Someone argues: [slippery slope argument]
Help me evaluate each step in the chain:
1. Is each step likely to follow from the previous?
2. What could prevent the slide?
3. Are there historical examples where this chain did or didn't happen?
Appeal to Emotion
What it is: Using emotional manipulation instead of logical reasoning.
Example: Showing a distressing image to argue for a policy, without addressing whether the policy would actually help.
Why it's tricky: Emotions aren't irrelevant to decisions. But emotions can be exploited. The question is whether the emotional appeal is attached to sound reasoning.
Red flags: When the emotional appeal seems designed to bypass thinking. When you're asked to feel rather than evaluate.
AI prompt:
Analyze this argument for emotional manipulation:
[Paste argument]
1. What emotions is it triggering?
2. Is there a logical argument underneath, or just emotional appeal?
3. Would I still agree if I weren't feeling this emotion?
Circular Reasoning (Begging the Question)
What it is: Assuming the thing you're trying to prove as part of your proof.
Example: "The Bible is true because it's the word of God, and we know it's the word of God because the Bible says so."
Why it's wrong: The conclusion is hidden in the premise. No new information is being provided.
Subtler forms: Sometimes circular reasoning is disguised by using different words. "We should trust experts because they have expertise" is circular — expertise and trustworthiness are the same claim restated.
Red Herring
What it is: Introducing an irrelevant topic to divert attention from the actual argument.
Example:
- Person A: "The company polluted the river."
- Person B: "But look at all the jobs they've created!"
Why it works: The diversion is often to something genuinely important, so you feel like you need to address it. But it's not relevant to the original claim.
AI prompt:
In this exchange, identify any red herrings:
[Paste conversation/argument]
What was the original point? What's being used to distract from it?
Appeal to Nature
What it is: Claiming something is good because it's "natural" or bad because it's "unnatural."
Example: "Organic food is healthier because it's natural."
Why it's wrong: Natural things can be harmful (arsenic, snake venom). Unnatural things can be beneficial (medicine, vaccines). "Natural" doesn't equal good.
Bandwagon Fallacy (Appeal to Popularity)
What it is: Claiming something is true or good because many people believe it.
Example: "Millions of people use this product, so it must be good."
Why it's wrong: Popular beliefs can be wrong. Truth isn't determined by vote.
The trap: Popularity isn't zero evidence — if many people find something useful, there might be something to it. The fallacy is treating popularity as proof.
Hasty Generalization
What it is: Drawing broad conclusions from insufficient evidence.
Example: "I met two rude people from that city. People there are rude."
Why it happens: Our brains are pattern-matching machines. We generalize automatically. The discipline is asking whether the sample size warrants the conclusion.
Using AI to Spot Fallacies
General Fallacy Check
Analyze this argument for logical fallacies:
[Paste argument]
For each fallacy you find:
1. Name the fallacy
2. Quote the specific part that commits it
3. Explain why it's fallacious
4. Suggest how the argument could be made without the fallacy
Check Your Own Arguments
I'm making this argument:
[Your argument]
Play devil's advocate. What fallacies might I be committing? Be tough — I want to strengthen my argument.
Fallacy Practice
Give me a paragraph that contains 3 different logical fallacies. Don't tell me which ones.
After I identify them, tell me if I was right.
The Meta-Fallacy: Fallacy Abuse
Here's an important warning: naming fallacies can become its own form of bad reasoning.
Fallacy fallacy: Assuming that because an argument contains a fallacy, its conclusion is false. Bad arguments can still have true conclusions.
Fallacy bullying: Using fallacy names to shut down conversation rather than engage with it. "That's ad hominem!" shouldn't end a discussion; it should redirect it.
Selective fallacy-hunting: Scrutinizing arguments you disagree with for fallacies while ignoring fallacies in arguments you like.
The goal isn't to "win" by catching fallacies. It's to reason more clearly. Apply the same rigor to your own arguments as to others'.
Fallacies in the Wild
Fallacies are everywhere once you learn to see them:
Political speeches: Heavy on emotional appeals, false dilemmas, and attacks on opponents' character.
Advertising: Appeals to authority (celebrity endorsements), bandwagon effects, false dilemmas ("you can't afford NOT to buy this").
Social media: Straw manning is epidemic. Everyone argues against the dumbest version of the other side.
Your own head: Motivated reasoning leads you to commit fallacies in favor of things you want to believe.
Practice exercise: For one day, note every fallacy you encounter. News, social media, conversations. You'll be shocked how common they are.
What Fallacies Don't Tell You
Identifying a fallacy tells you the argument is flawed. It doesn't tell you the conclusion is wrong.
"We should eat healthy because a celebrity said so" is a bad argument. But eating healthy might still be a good idea.
After identifying a fallacy, ask: "Is there a better argument for this conclusion?" Often there is. The person just made a bad argument for a potentially good conclusion.
This keeps you from using fallacy-spotting as a way to dismiss ideas you don't like.
What's Next
Logical fallacies are errors in argument structure. But even sound arguments can lead you astray if you're feeding them biased inputs.
Chapter 3 covers cognitive biases — the systematic ways your brain distorts perception and judgment before you even start reasoning.