ProjekinAja. Clear Thinking Stage 1 · Core Reasoning Lesson 1.4
● Stage 1 · Lesson 1.4

Informal Fallacies by Family

Thirteen reasoning errors grouped into four families — and why spotting a fallacy is only the beginning of a real rebuttal.

⏱ ~18 min read 🧩 Core skill 🎮 Quiz included

🎯 By the end you'll be able to

Place any informal fallacy into its OpenStax family, identify which premise is defective and why, cite the evidence or counterexample that shows it, and state whether the defect destroys or merely weakens the conclusion.

RUNNING EXAMPLEThe argument we'll keep testing

Throughout this lesson we'll evaluate a single argument so you can see different fallacies at work on the same material:

P1: "It is wrong to profit when another person loses."
P2: "Whenever someone profits from investment, another person loses."
C: "Therefore, investing is wrong."

This argument is structurally valid — if both premises were true, the conclusion would follow. The question is whether the premises hold up. You'll see multiple fallacy families attacking this argument from different angles as we go.

OVERVIEWThe four families

OpenStax groups informal fallacies into four families based on where the argument goes wrong. Knowing the family tells you what question to ask:

FamilyCore defectKey question to ask
1 · Relevance A premise doesn't actually support the conclusion "Does this premise have any bearing on the claim?"
2 · Weak induction Some connection exists, but evidence is too thin to carry the weight placed on it "Is the evidence strong enough and representative enough?"
3 · Unwarranted assumptions A premise quietly smuggles in something unestablished "What is this premise assuming that it hasn't proved?"
4 · Diversion Attention is redirected away from the real issue "Is the actual question still being addressed?"
🔑 The payoff of a taxonomy

Once you can name the family, you know the repair strategy. Relevance errors require showing the missing connection. Weak-induction errors require better evidence. Unwarranted assumptions require proving the assumption. Diversion errors require returning to the original question.

FAMILY 1Relevance errors — the premise doesn't connect

In a relevance fallacy, a premise is offered as support for the conclusion but has no genuine logical bearing on it. The argument can feel compelling because the premise may be true — it just doesn't help establish the conclusion.

Ad hominem / personal attack

📖 Definition

Attacking the person making the argument rather than the argument itself. The person's character, motives, or circumstances are irrelevant to whether their premises are true.

"Why should we trust her argument that investing creates wealth? She works for a brokerage — of course she'd say that."

Her employment is a fact about her, not about the argument. Even if she has self-interested motives, the argument's premises must be evaluated on their own merits. (A conflict of interest is a reason to scrutinise evidence more carefully — it's not a rebuttal.)

Appeal to popularity

📖 Definition

Claiming something is true (or right) because many people believe it. Also called ad populum. Popularity is evidence of belief, not of truth.

"Millions of people believe investing is morally wrong because it exploits others. Surely that many people can't be mistaken."

They can. Widespread belief in flat earth, geocentrism, and bloodletting as medicine are three historical reminders. The question is whether P2 is true — not how many people accept it.

Appeal to emotion

📖 Definition

Using emotional language or imagery as a substitute for evidence, rather than as an accurate signal alongside it. Emotion can be genuinely relevant — if a policy causes suffering, that suffering matters. The fallacy is when emotion does the work that premises should do.

"Think of the retiree who lost half her savings when the market crashed. How can you defend the system that did that to her?"

The story is poignant and real — but it doesn't establish P2. Markets decline for many reasons; that decline doesn't prove that every investor's gain required someone else's loss.

FAMILY 2Weak induction — evidence too thin to carry the load

Here the premise does have some connection to the conclusion — but the inductive leap is too large for the evidence offered. The argument isn't completely irrelevant; it's just dramatically under-supported.

Hasty generalization

📖 Definition

Drawing a broad conclusion from a sample that is too small, cherry-picked, or unrepresentative of the target population.

"I've watched three people lose money in the stock market while their brokers got rich. Clearly investing is a zero-sum system."

Three observations — selected because they were memorable — don't establish P2 for the full scope of all investment activity across all markets. The sample is both tiny and biased toward failure stories.

Weak analogy

📖 Definition

Arguing from a similarity between two things to a conclusion that requires the similarity to hold in the relevant dimension — when it doesn't.

"Investing is like gambling: you put money in, you might win, someone else might lose. So investing is as morally questionable as gambling."

Investing and gambling share risk and uncertain outcomes. But they differ in the relevant dimension: gambling is a designed zero-sum system (the house always has an edge); investing in productive businesses can generate new value that didn't exist before, producing non-zero-sum outcomes. The analogy breaks precisely where the argument needs it to hold.

Unrepresentative sample

📖 Definition

Similar to hasty generalization, but emphasizing that the sample is skewed in a particular direction rather than merely small — the conclusion is extrapolated from data that systematically misrepresents the population.

"In surveys of people who lost money in the 2008 crash, almost everyone said investing hurt them. Investing clearly destroys wealth."

Surveying only crash-year losers excludes the much larger population of long-term investors who recovered and grew their wealth. The sample's selection mechanism guarantees a distorted result.

Unreliable authority

📖 Definition

Citing an authority whose expertise doesn't cover the claim being made, or whose expertise is genuine but whose testimony on this particular point is contested or self-interested.

"Professor X, one of the world's leading philosophers, says investing is morally indefensible. So we should accept that it is."

If Professor X's specialty is medieval ethics and they haven't engaged with modern financial economics, their view on whether markets are zero-sum carries little evidential weight on the empirical question in P2. Authority is strong evidence within its domain and weaker outside it.

💡 Legitimate appeals to authority

Not all appeals to authority are weak. If a climate scientist says average global temperatures have risen, that's appropriate evidence. The question is: is this person's expertise directly relevant to this specific claim, and is their view representative of the expert community rather than an outlier?

FAMILY 3Unwarranted assumptions — smuggling in the unestablished

These fallacies don't openly attack the conclusion — they quietly assume something that hasn't been established, building the argument on a foundation that looks solid but isn't there. They're especially important for the running example.

False dilemma

📖 Definition

Presenting a choice as if only two options exist when the actual option space is larger. Also called false dichotomy.

"Either you believe investing is moral despite the harm it causes, or you accept that the entire financial system is corrupt and must be dismantled."

This ignores a large range of positions: that some forms of investment create value while others don't, that regulation can address harms without dismantling the system, that the moral status of investing depends heavily on context. The binary collapses a spectrum.

Begging the question

📖 Definition

Assuming in a premise the very conclusion you're trying to prove. The argument is circular — it contains the conclusion as a hidden input.

"Investing is wrong because it's the kind of activity that causes unjust harm — and anything that causes unjust harm is wrong."

"Unjust harm" is the conclusion in disguise. The argument assumes what investing does counts as "unjust harm," which is exactly the proposition being contested.

Loaded question

📖 Definition

Embedding an unestablished assumption inside a question so that any direct answer appears to concede the assumption.

"When did you stop feeling guilty about profiting at other people's expense?"

The question assumes the listener profits at others' expense and has felt guilty about it. Answering "I haven't stopped" or "recently" both concede the assumption. The only honest response is to challenge the embedded premise.

Oversimplifying complexity

📖 Definition

Reducing a genuinely complicated causal or conceptual picture to a simple mechanism, discarding the complexity that changes the conclusion.

"Markets are just people trading — and in every trade, there's a winner and a loser. So markets are zero-sum by definition."

This ignores value creation: a company that invents a medicine creates real value that benefits both buyers (who get health) and sellers (who get revenue). Trade isn't merely redistribution — it can generate surplus. The simplification erases the most important part of the picture.

Assuming all gains require losses — the running example's core flaw

📖 Definition

Treating economic exchanges as necessarily zero-sum when many (especially productive investment) are positive-sum: both parties can be better off, and new value can be created rather than merely transferred.

This is the defect in P2 of the running example. P2 says "whenever someone profits from investment, another person loses." This is silently assuming a closed-pool, zero-sum model of all investment activity.

The assumption may hold for some financial instruments — a zero-sum futures contract or a pure speculation trade on a fixed-supply asset. But it fails for many others: investing in a startup that creates new employment and products, buying shares in a company that grows its earnings, or lending capital that enables a business to produce more than existed before. In these cases, investor profit can coexist with real gains for workers, customers, and the wider economy.

⚠ Why this matters for the running argument

P2 is the load-bearing premise of the running argument. If P2 is false — if not all investment profit requires someone else's loss — then the argument fails. The conclusion might still be partially correct (some forms of speculation may be zero-sum or worse), but the sweeping claim that all investing is wrong is not established. We'll examine the scope of P2 in much more detail in the next lesson.

FAMILY 4Diversion — redirecting attention

Diversion fallacies don't directly attack the argument — they move the conversation somewhere else so the original question never gets answered.

Red herring

📖 Definition

Introducing a tangential issue that is related enough to sound relevant but doesn't bear on the actual question. Named after the practice of dragging a smelly fish across a trail to confuse hunting dogs.

"Before we debate whether investing is morally wrong, consider that half the world's population lives on less than $6.85 a day while investors accumulate enormous wealth."

Global poverty is real and important. But the question is whether the logical structure of investment is inherently zero-sum. Introducing wealth inequality shifts the conversation to a different — if related — topic. The original premises remain unexamined.

Straw man

📖 Definition

Misrepresenting an opponent's actual position — usually making it weaker, more extreme, or more absurd than it is — and then attacking the distorted version.

"The pro-investment side just wants to let the wealthy exploit everyone else without limit. They think greed is good and the poor deserve their fate."

This takes a reasonable position ("investing can create value and be beneficial") and replaces it with a cartoon villain. Defeating the cartoon doesn't engage the actual argument. The honest response to an argument is to represent it accurately and then critique it.

💡 Steelmanning as the antidote

The opposite of a straw man is a steel man: represent the opposing argument in its strongest, most charitable form, then critique it. This is harder but produces better thinking. If you can only defeat the weak version of an argument, you haven't actually resolved the question.

INTERACTIVEFallacy spotter — test yourself

Six statements connected to the running example and related topics. Identify the fallacy family and specific fallacy. Click your answer for immediate feedback.

Each question is independent — you can revisit them. Score appears when all six are answered.

🎯 Interactive · Fallacy Spotter by Family

1. "You can't critique the investment industry — you have a pension fund yourself. That makes you a hypocrite and your argument worthless."

2. "Most people I know feel that making money from money while others struggle is just wrong. That widespread intuition must mean something."

3. "I invested in three different start-ups and all three failed, taking investors' money with them. Investment is clearly a system that destroys value."

4. "Either you accept that all profit from investment is morally clean, or you must admit the whole capitalist system is exploitation. Which is it?"

5. "The people who defend investing just want to protect the rich at the expense of the working class. They don't care about ordinary people."

6. "When did you realise that investing is just legal theft from people who can't afford to play the game?"

CRITICAL WARNINGNaming a fallacy does NOT refute an argument

⚠ The most important thing in this lesson

Labelling a fallacy is a starting point, not a finishing move. It tells you where to look — it does not tell you whether the conclusion is false.

Suppose someone commits a hasty generalization when arguing that investing is wrong. Even if the argument is fallacious, the conclusion might still be correct — you'd just need a better argument to establish it. Conversely, an argument free of fallacies might still lead to a false conclusion if the premises happen to be false.

A complete rebuttal of a fallacious argument requires four things:

StepWhat you must sayApplied to P2 of the running example
1. Name the defective premise Which specific premise carries the flaw? "Premise 2 — 'whenever someone profits from investment, another person loses' — is the defective premise."
2. Identify the defect What kind of error? Which family? "This falls under unwarranted assumptions — it silently assumes all investment is zero-sum without establishing that."
3. Provide evidence or a counterexample What shows the premise fails? "A startup that employs 50 people and sells a useful product creates new value. The investor's profit coexists with value gained by employees, customers, and the wider economy — not taken from them."
4. State the consequence for the conclusion Does this destroy or merely weaken the conclusion? "This destroys the universal conclusion. Investing as a whole is not shown to be wrong. A narrower conclusion — that some zero-sum speculative instruments may be morally problematic — remains open."

Notice step 4 is important: sometimes a fallacious premise weakens rather than destroys the conclusion. If the conclusion was modest ("investing sometimes harms others"), a hasty generalization doesn't fully refute it — it just fails to establish it. If the conclusion was sweeping ("investing is always wrong"), the same fallacy is more damaging.

CONTEXTInformal logic is bigger than fallacy-hunting

It's tempting, having learned fourteen fallacy names, to go looking for them everywhere. Resist this. Fallacy identification is one tool in informal logic — the study of real-world argument as it actually appears in law courts, medical consultations, political debates, business decisions, and everyday conversation.

Informal logic also includes:

A skilled reasoner uses fallacy detection as one entry point into a harder question: what would it take to actually establish or defeat this conclusion? That question is always more important than the fallacy label itself.

💡 In practice: law, medicine, politics

In courtrooms, lawyers must distinguish admissible evidence from rhetorical distraction — the rules of evidence are partly a formal system for excluding relevance errors and weak inductions. In medicine, clinicians must assess whether a causal claim about a treatment has adequate evidential support or is based on an unrepresentative sample. In political argument, all four families appear daily. The point of this lesson is to make you usefully suspicious about how conclusions are being established, not just what they are.

WEAKNESSCommon mistakes when studying fallacies

✗ "I named the fallacy, therefore the argument is refuted."

Naming shows where the problem is. Refuting requires showing why that specific premise fails and what that means for the conclusion. The four-step framework above is the minimum for a real rebuttal.

✗ "Any appeal to emotion is automatically a fallacy."

Emotional considerations can be genuinely relevant. If a policy causes suffering, that suffering is real evidence. The fallacy is when emotion substitutes for evidence — not when it accurately signals something that matters.

✗ Treating all arguments from authority as fallacious.

Expert testimony is legitimate evidence within its domain. "Unreliable authority" is a fallacy only when the authority lacks relevant expertise, has an undisclosed conflict of interest that distorts their view, or when their opinion is cited as if it settles an empirical question that requires direct evidence.

✗ Assuming every "if A then B then C" chain is a slippery slope.

A genuine causal chain argument with evidence for each step is not a fallacy. The slope is slippery only when the links are asserted without evidence and presented as inevitable rather than merely possible.

PRACTICE🔨 Weekly exercise

🔨 This week's exercise

Find one argument in the wild — a news article, a social media post, a conversation, a podcast — that you suspect contains a fallacy. Then do the full four-step analysis:

  1. Reconstruct the argument: what are the premises, what is the conclusion?
  2. Identify which premise is defective and which family the error belongs to.
  3. Write down the evidence or counterexample that shows why the premise fails.
  4. State whether the defect destroys the conclusion entirely, or only weakens the strongest version of it.

One complete analysis is worth more than ten fallacy labels. The habit of asking "what would it actually take to establish this conclusion?" is what separates careful reasoning from fallacy-hunting.

GLOSSARYKey terms introduced

Relevance fallacy
A premise offered in support of a conclusion that has no genuine logical bearing on it.
Ad hominem
Attacking the person making an argument rather than the argument's premises.
Appeal to popularity (ad populum)
Claiming a conclusion is true because many people believe it.
Appeal to emotion
Using emotional language or imagery as a substitute for evidence rather than alongside it.
Red herring
Introducing an irrelevant topic that distracts from the original question.
Weak induction
An argument where premises have some connection to the conclusion but the evidential support is too thin to carry the weight placed on it.
Hasty generalization
A broad conclusion drawn from too small or unrepresentative a sample.
Weak analogy
Arguing from a similarity that doesn't hold in the dimension the argument requires.
Unwarranted assumption
A premise that quietly assumes something unestablished, building the argument on an invisible foundation.
False dilemma
Presenting only two options when the actual option space is larger.
Begging the question
Assuming in a premise the very conclusion the argument is meant to establish.
Loaded question
Embedding an unestablished assumption inside a question so any direct answer appears to concede it.
Zero-sum assumption
Treating an economic transaction as necessarily having a winner and a loser, when positive-sum outcomes (value creation) are possible.
Diversion
A family of fallacies that redirect attention away from the real question rather than engaging it.
Straw man
Misrepresenting an opponent's position to make it weaker, then attacking the misrepresentation.
Steel man
The opposite of straw man — representing the opposing argument in its strongest form before critiquing it.

ACTIVE RECALLCheck your understanding

Answer in your own words first, then reveal.

1Name the four fallacy families and the core defect each represents.›

Relevance: a premise doesn't have genuine bearing on the conclusion. Weak induction: the premise has some connection to the conclusion but evidence is too thin. Unwarranted assumptions: a premise quietly assumes something unestablished. Diversion: attention is redirected away from the actual question.

2What four steps are required to fully rebut a fallacious argument — not just name the fallacy?›

(1) Name the defective premise. (2) Identify the defect (fallacy family and specific type). (3) Provide evidence or a counterexample showing why the premise fails. (4) State whether the defect destroys the conclusion entirely or merely weakens it.

3Why does P2 of the running example commit an unwarranted assumption? What counterexample undermines it?›

P2 silently assumes all investment is zero-sum — that every gain requires someone else's loss. This fails for productive investment: a business that creates new goods, services, or employment generates value that didn't exist before. Both investors and the wider economy can benefit. A startup that grows from seed funding into a company with employees and customers is a concrete counterexample: the investor's return coexists with new value for workers and customers, not at their expense.

4What is the difference between a straw man and a steel man? Why does the distinction matter?›

A straw man misrepresents an opponent's argument in a weaker or more extreme form and attacks that. A steel man represents it in the strongest, most charitable form and then critiques it. The distinction matters because defeating a straw man doesn't advance understanding — it produces the illusion of refutation without engaging the real argument. Steel-manning is the honest practice.

Progress saves in this browser. Next: how the scope of words like "all" and "never" changes everything →