Skip to content

Note · Fundamental concepts

What is a logical fallacy? How arguments go wrong

A logical fallacy is an error in reasoning that makes an argument seem stronger than it is. The formal and informal kinds, the common ones with examples, and how to tell a fallacy from a fair point.

This note and the comments marked the Graphite mark are Graphite’s: an AI model, Claude Opus 5.5, wrote them on 5 November 2026, following Graphite’s own instructions, from the sources at the note’s end, and checked them against those sources. Check them as you would anyone’s.

This note and its comments are written by an AI model.

Contents · 7 sections

A logical fallacy is an error in reasoning that makes an argument look stronger than it really is: the premises seem to support the conclusion, but they don't. Some fallacies are errors of form, a broken link that no subject matter can fix. Most are informal: they attack the wrong target, assume what they claim to prove, or hide a weak step. Spotting one shows that the argument fails, not that its conclusion is false.

In short

  • A fallacy is reasoning that seems to support its conclusion but doesn't. Formal fallacies break the shape of the argument; informal ones go wrong in its content or context.
  • Many named fallacies have a fair twin: appealing to an expert, warning of a likely chain of events or questioning a source's interests can all be reasonable.
  • A fallacy shows that an argument doesn't prove its point. Concluding that the point is false is a fallacy of its own.

What makes an argument a fallacy?

An argument is a set of premises offered in support of a conclusion. It goes wrong in one of two ways: a premise is false, or the premises, even if true, don't support the conclusion. A fallacy is the second kind of failure dressed up as success. The Stanford Encyclopedia of Philosophy takes as its working definition "an argument that seems to be better than it really is", and dates the modern study of fallacies to the logician Charles Hamblin's book Fallacies of 1970.

The study is old. Aristotle's On Sophistical Refutations, from the 4th century BCE, describes 13 ways an argument can only seem to refute an opponent, six that depend on tricks of language and seven that don't. Many fallacy names are Latin because later scholars cataloged them that way: ad hominem, "to the person"; post hoc ergo propter hoc, "after this, therefore because of this".

What is the difference between formal and informal fallacies?

A formal fallacy is an argument whose form is invalid: true premises of that shape can lead to a false conclusion, whatever the subject. You can expose one by keeping the form and changing the content, the test described in what logic is.

Formal fallacy Form Example Why it fails
Affirming the consequent If P then Q. Q. So P. If the oven is on, the kitchen is warm. The kitchen is warm. So the oven is on. The heating, the sun or a crowd could warm the kitchen
Denying the antecedent If P then Q. Not P. So not Q. If it's a weekend, the bank is closed. It's not a weekend. So the bank is open. Holidays and closing time exist
Undistributed middle All A are B. All C are B. So all A are C. All cats are mammals. All dogs are mammals. So all cats are dogs. Sharing a trait doesn't make two groups the same

An informal fallacy has nothing wrong with its shape. It fails because of what it says: it changes the subject, relies on a word used two ways, or leans on a premise that is weak or assumed. Most of the fallacies people name in everyday arguments are informal.

What are the most common informal fallacies?

Each of these is a real error, but many have a fair version that looks similar. The last column says how to tell them apart, which matters more than the name.

Fallacy What it does Everyday example When a similar move is fair
Straw man Answers a weaker or distorted version of the other side's view "You want the library open on Sundays? So you think staff should never get a day off." Restating someone's view and asking if you've got it right
Ad hominem Attacks the person instead of the argument "Don't listen to her budget plan; she can't even keep her garden tidy." Questioning a witness's reliability or a source's financial interest, when the claim rests on their word
False dilemma Offers two options when there are more "Either we buy the expensive printer or we go without one." When there really are only two options, and that has been shown
Slippery slope Claims one step will lead to an extreme, without showing each link "If we let one student hand in an essay late, soon nobody will meet any deadline." When each step in the chain is shown to be likely
Appeal to authority Treats someone's word as proof, especially outside their expertise "A famous chef says this sunscreen doesn't work." Relying on the consensus of experts in their own field
Begging the question Assumes the conclusion in a premise "This guidebook is reliable because it says it's the most accurate guide." Never: an argument that assumes its conclusion has shown nothing
Equivocation Uses one word in two senses "A feather is light. What is light can't be dark. So a feather can't be dark." Never in the same argument; clarify which sense is meant
Post hoc Assumes that because B followed A, A caused B "I started taking zinc and my cold went away in a week." When other causes have been ruled out, ideally by a trial
Appeal to ignorance Says a claim is true because it hasn't been disproved "No one has proved the old well isn't haunted, so it is." When a thorough search would have found the thing if it existed

Post hoc is the everyday form of confusing correlation with causation: most colds clear up within one to two weeks whatever you take. The sunk cost fallacy stretches the word from arguments to decisions: the money already spent is offered as a reason for a choice that it doesn't support.

How do you check an argument for fallacies?

Here is a sales pitch worked through. "Either you sign the two-year gym contract today or you'll never get fit. Everyone serious about fitness signs one; a gold-medal swimmer recommends our club. And people who pay monthly always quit; my cousin did."

  1. Find the conclusion. Sign the two-year contract today.
  2. List the reasons. Only two options exist; everyone serious signs; a champion recommends it; monthly payers quit.
  3. Check each link, not just each fact. "Today or never" is a false dilemma: you could sign next week, pay monthly or exercise at home. "Everyone serious signs" appeals to popularity, and isn't shown. The swimmer may be an expert on swimming but is likely paid for the endorsement, so their word is weak evidence about this contract. "My cousin quit" is a single case standing in for a rule, a hasty generalization.
  4. Ask what's left. None of the reasons, as given, supports the conclusion.
  5. Don't overshoot. The contract might still be good value. The pitch failed to show it; you'd need the price, the terms and the cancellation rules to decide.

A fallacy is a reason to reject an argument, not a reason to reject its conclusion. Concluding "the argument was fallacious, so the claim is false" is itself a fallacy, known as the argument from fallacy or the fallacy fallacy. A true claim can be argued for badly.

How are fallacies different from biases?

A fallacy is a flaw in an argument, something you can point to on the page. A cognitive bias is a tilt in the person doing the judging, such as noticing only the evidence that fits what you already believe. Biases often make fallacies feel convincing, and the two are easy to blur, but the fix differs: a fallacy is answered by pointing out the missing link, a bias by procedures that check your own judgment.

Still argued

Philosophers disagree about how useful fallacy lists are. Some scholars of argument, such as Douglas Walton, argue that many "fallacies" are reasonable moves in some contexts and faulty in others, so a list of names misleads: what matters is whether the move fits the conversation. Others defend the lists as a practical first check. Both sides warn against fallacy-spotting as a debating weapon, naming a fallacy instead of engaging with the point.

Where does this lead next?

Fallacies are the failures of evidence and logic in practice. The habits that catch them, asking what supports a claim and what else could explain it, are the core of critical thinking.

Check yourself

  • Without looking back, explain the difference between a formal and an informal fallacy, with one example of each.
  • When is relying on an expert reasonable, and when does it become a fallacy?
  • Why doesn't showing that an argument is fallacious prove its conclusion is false?

Sources

Published by Graphite on , set the way Graphite publishes a note.

This note and the comments marked the Graphite mark are Graphite’s: an AI model, Claude Opus 5.5, wrote them on 5 November 2026, following Graphite’s own instructions, from the sources at the note’s end, and checked them against those sources. Check them as you would anyone’s.

About the author

Graphite

An AI model, Claude Opus 5.5, wrote it for Graphite from the sources at its end.

More in Fundamental conceptsSearch the MarketplaceWho writes here