Logic and Philosophy · Unit 2 · High school
Informal Fallacies: Spotting Bad Reasoning
A weak argument can still sound finished. Here's how to check.
In Unit 1: Arguments you learned to find premises and a conclusion, test validity with a counterexample, and name two formal errors: affirming the consequent and denying the antecedent. Those mistakes live in the pattern itself. Change the topic and the pattern is still broken.
This unit is about a different family of errors: informal fallacies. They depend on the content and the context, not just the shape of the argument. The same pattern can be fair in one case and faulty in another. You'll learn seven of the most useful labels, when an appeal is not fallacious, and how to answer a bad argument without making one of your own.
What a fallacy is
A fallacy is a common error in reasoning: an argument that seems better than it really is. The first known systematic study of these errors is Aristotle's Sophistical Refutations (De Sophisticis Elenchis), written in the fourth century BCE. Many of the Latin names you'll meet later come from the long life of that tradition.
Logicians usually split fallacies into two groups:
- Formal fallacies fail because of their logical form. You met two in Unit 1: affirming the consequent and denying the antecedent. If the pattern is invalid, the topic doesn't save it.
- Informal fallacies depend on the content, the context, or both. You often cannot spot them by looking only at the pattern. A slippery-slope shape, for example, appears in both weak scare stories and careful risk arguments. What matters is whether the steps in between are actually supported.
A charge of fallacy always needs a reason. Naming the pattern is a start, not a proof. If you claim someone's reasoning is fallacious, the burden is on you to show the fault.
Your teacher on this course is Socrates, a Jugami character inspired by the Athenian philosopher who became famous for testing whether a claim was as strong as it sounded. The aim here is the same: not to collect labels, but to see whether the reasons really support the conclusion. This course does not invent lines for the historical Socrates to have spoken.
Seven informal fallacies
These seven show up constantly in class discussion, comments, and public debate. For each one, notice the companion lesson: the same move is sometimes fair.
| Fallacy | The mistake | When a similar move can be fair |
|---|---|---|
| Ad hominem | Attacking the person instead of their reasons | A fact about the speaker is actually relevant to the issue, such as a conflict of interest in testimony |
| Straw man | Refuting a distorted, weaker version of the view | An accurate restatement, even a sharp one, of what was really claimed |
| False dilemma | Offering too few choices and demanding one of them | The options really do exhaust the genuine possibilities |
| Slippery slope | Treating a chain of later harms as inevitable without support | Each step is backed by evidence, so the risk is real |
| Appeal to authority | Treating an unqualified or off-topic voice as if it settled the question | A relevant expert, speaking in their field, with a checkable basis |
| Hasty generalization | Leaping from too few, or unrepresentative, cases to a wide claim | A modest conclusion that fits the size and quality of the evidence |
| Begging the question | Using a premise that already assumes the conclusion | Giving an independent reason that can be assessed without already accepting the conclusion |
Ad hominem
Ad hominem means "to the person." The fallacy is an irrelevant attack on the speaker, offered as if it undermined the argument itself.
- Andre proposed a new recycling schedule and gave reasons about cost and pickup times.
- Andre failed chemistry last year.
- So Andre's recycling plan should be rejected.
The plan might still be a good one. A chemistry grade does not answer the reasons about cost and pickup. The argument should stand or fall on those reasons.
When it is not a fallacy. Sometimes a fact about the person is relevant. If Andre is testifying that he saw something happen, his honesty or a conflict of interest matters to how much weight you give the testimony. If you are hiring a chemistry tutor, last year's grade is on topic. Even then, a conflict of interest is a reason to check the evidence more carefully, not automatic proof that the conclusion is false. The evidence might still be good.
A close cousin is tu quoque ("you too"): rejecting advice because the speaker doesn't follow it. "You told me to exercise, but you don't run, so I can ignore the advice." The value of the advice is not settled by the advisor's habits.
Straw man
A straw man distorts someone's view into a weaker or more extreme claim, refutes that version, and treats the original view as defeated.
Leah: "The library should stay open one extra hour on weeknights so more students can finish research."
Omar: "Leah wants the school open all night. That's unsafe and expensive, so we should ignore her idea."
Omar has an easier target, but it is not Leah's target. She asked for one hour, not an all-night building.
When it is not a fallacy. Restating a view accurately, even in blunt language, is not a straw man. The fault is the distortion. If you can put the other person's claim in words they would accept, you are on safer ground.
False dilemma
A false dilemma (also called a false dichotomy, or a black-or-white fallacy when only two choices are offered) shrinks the options unfairly and then demands that you pick from that short menu.
"Either we cancel the spring play, or we spend unlimited money on new sets. We can't spend unlimited money. So we have to cancel the play."
Those are not the only options. The drama club could reuse last year's flats, borrow costumes, or scale the set down.
When it is not a fallacy. Some menus really are complete, or complete enough for the decision at hand. "You can accept this job offer or decline it." "This motion either passes or it doesn't." The test is whether the limited list is fair, not whether it is short.
Slippery slope
A slippery slope argument says that if we take a first step, a chain of further steps will follow until we reach an unacceptable result, so we should reject the first step. The form itself is not the problem. The quality of the argument depends on how strong the links are from one step to the next.
"If we allow hats in class, soon there will be no dress code at all, then students will wear anything, and the school will lose all order. So no hats."
The first step is modest. The later disasters are only assumed. Nothing in the argument shows that hats make a collapsed dress code likely.
When it is not a fallacy. If each step is backed by evidence, the warning can be fair. A chemistry teacher who says "If you skip the burner's safety check, a gas leak can ignite, and that can start a fire" is not inventing a panic story. There is a known mechanism from one step to the next. Ask for the evidence in the middle, not just for a label.
Appeal to authority
An appeal to authority (in Latin, argumentum ad verecundiam) backs a claim by pointing to what some authority says. Most reasoning of this kind is not fallacious. A large part of what we justifiably believe comes from people who know more than we do about a subject. The appeal becomes fallacious when the person is not an authority on this question, when they cannot be trusted to tell the truth about it, when qualified experts are divided and one voice is treated as decisive, or when the authority is misquoted.
"This brand of graphing calculator is the best one for calculus, because a famous sprinter said so in an ad."
The sprinter may be an authority on sprinting. That does not make them an authority on calculators or calculus.
When it is not a fallacy. "Our calculus teacher, walking through a worked method in the course manual, says this is the standard way to check the derivative on this model of calculator." The speaker is a relevant expert, in their field, offering a checkable basis. That is how appeals to authority are supposed to work.
Hasty generalization
A hasty generalization leaps to a broad conclusion from too little evidence: a sample that is too small, or too unrepresentative, for the claim being made.
"The two people I know in the chess club are quiet, so chess players are all shy."
Two acquaintances cannot support a claim about an entire group. A vivid personal story can be especially tempting here, because it feels like firsthand knowledge. It is still only two cases.
When it is not a fallacy. A modest conclusion can fit modest evidence. "This printer jammed the last three times I used it this week, so I'm going to try a different printer today" does not claim that every printer is hopeless. Match the size of the conclusion to the size of the evidence.
Begging the question
Begging the question (in Latin, petitio principii) is circular reasoning: the conclusion is derived from premises that already assume the conclusion. In everyday talk, people sometimes use "that begs the question" to mean "that raises a further question." In logic, the phrase is narrower. The argument goes in a circle instead of moving to a new reason.
"The dress code is reasonable, because it is a sensible policy for our school."
"Reasonable" and "sensible" are doing the same work. Anyone who doubts the conclusion already doubts the premise. Notice a surprising point: an argument can beg the question and still be valid. If the premise is true, the conclusion follows. The problem is that the argument doesn't give you an independent reason to accept the premise in the first place. It doesn't make progress.
When it is not a fallacy. Offer a reason that can be checked without already granting the conclusion: "The dress code is reasonable because it was written with student input, it is applied consistently, and it still allows students to dress for religious or medical needs." Those are claims you can investigate even if you start out unsure.
How to answer a fallacy fairly
Spotting a fallacy is only half the work. The other half is answering it without committing one yourself.
- Restate the actual claim charitably. Put it in words the other person would accept. That is the opposite of a straw man.
- If you call it a fallacy, justify the charge. The burden is on you. Point to the specific gap: the missing option, the weak link in the chain, the irrelevant personal remark, the sample that is too small, or the premise that restates the conclusion.
- Ask for what is missing, rather than only applying a label. "What is the evidence that hats lead to no dress code?" is more useful than "That's a slippery slope."
- Don't answer a fallacy with a fallacy. Insulting the speaker, exaggerating their view, or saying "you do it too" repeats the problem.
- Be ready to withdraw the charge. If the person turns out to be a relevant expert, or the slope has evidence, or the dilemma is genuine, the fair move is to say so and go back to the reasons.
A short model, using Leah's library proposal from above:
"Leah asked for one extra weeknight hour, not an all-night building. If the worry is safety or cost for that single hour, we should look at staffing and the actual budget. Those are fair questions. They don't require treating her proposal as something she didn't say."
Practice problems
Printable worksheet: Spotting informal fallacies (PDF)
The worksheet has all-new arguments to label, repair and judge. None of them appear on this page or in the self-guided quiz, so it's a real check of what you've learned.
Self-guided quiz
Eight questions. Choose an answer to see feedback right away. Your first pick locks in and is the one that counts. Every answer explains why it's right or wrong and which section to reread.
Talk about it
- When is it fair to bring up a speaker's conflict of interest, and when does that become a way to avoid their evidence?
- Have you ever realized you were arguing against a weaker version of someone's view than they actually held? What changed when you restated it more carefully?
- Why might "that's a fallacy" be a less useful reply than naming the missing option, the weak link, or the too-small sample?
- People sometimes say "that begs the question" when they mean "that raises a further question." Does the mix-up matter in a serious discussion?
Sources
- What a fallacy is; formal vs. informal fallacies; Aristotle's Sophistical Refutations as the first known systematic study; the burden of justifying a fallacy charge; and the entries used below: Internet Encyclopedia of Philosophy, "Fallacies"
- Informal fallacies as arguments that seem better than they are; straw man as a distortion of an opponent's view; ad hominem (abusive, circumstantial, and tu quoque); appeal to authority; begging the question as circular and not necessarily invalid; slippery slope: Stanford Encyclopedia of Philosophy, "Fallacies"
- Ad hominem, including when a personal fact can be relevant; straw man / caricaturization; false dilemma and black-or-white (the limited menu must be unfair); slippery slope (the form appears in both good and bad arguments; quality depends on the probabilities of the steps); appeal to authority (most such appeals are not fallacious; they fail when the authority is unqualified, untrustworthy, divided, or misquoted); hasty generalization from too small a sample; begging the question / petitio principii: Internet Encyclopedia of Philosophy, "Fallacies", entries under those names
- Validity, soundness and formal pattern, for the link back to Unit 1: Internet Encyclopedia of Philosophy, "Validity and Soundness"; Arguments: Premises, Conclusions and Counterexamples
The examples on this page are original illustrations of those patterns. They are not taken from the sources, and they do not repeat the self-guided quiz or the worksheet.
Credits: Art: all drawings are original, hand-coded SVG made for Jugami; the share images are rendered from the same art. Our teacher Socrates is a character inspired by the Athenian philosopher, not a real person speaking.