Logic and Scientific Thinking

Informal Fallacies 2

Bogdan G. Popescu

Tecnologico de Monterrey

Act III: Distortion and Emotional Appeals

Recap

Question	Answer from Session 7
The three-step diagnosis?	Name the move · quote the words · say what would decide it
Faulty reasoning?	False cause · circular reasoning
Misused sources?	Ad hominem · appeal to authority

Two families down — two to go: distorted claims and emotional appeals.

Today’s Route

Today: distorted claims and emotional appeals — plus one neighbor (the verbal dispute) and the volume tactic.

Faulty reasoningMisused sourcesDistorted claimsEmotional appeals

1false cause

2circular reasoning

3ad hominem

4appeal to authority

5straw man

6red herring

7slippery slope

8equivocation

9false dilemma

10appeal to fear

11appeal to pity

12bandwagon

∞Gish Gallopthe volume tactic — carries any move above; felt previouslymeta

Straw Man

“S1: I’d rather not get a dog.”
“S2: So you hate animals?”

S2 rebuilt S1’s claim
S2 likely wants to attack S1
Test: would the opponent accept your summary of their view?

Lead with the domestic version — everyone has lived this argument, and the gap between “I’d rather not get a dog” and “you hate animals” is instantly visible (and it keeps us in the dog world from the formal session). Then run the same move VERBALLY on the running curfew case: critics said “the curfew may not cut crime”; the reply rebuilt it as “opponents want criminals to run free” — what critics actually said, then the manufactured copy. The distortion IS the fallacy — everything after it is theater; the original claim is never touched. The clip next slide performs the same move live. Verbal aside if time allows (NOT exam material): the cousin “nutpicking” — debating your rival’s worst spokesman instead of their best case. One label fewer on the slide; the room is already carrying eight. SCHOPENHAUER’S VERSIONS (assigned reading; Saunders 1896): Stratagem I is the straw man by extension — stretch the claim “beyond its natural limits” and attack the stretched copy (his own defense: “an accurate statement of the point at issue,” p. 15). Stratagem XXIII is the manufactured straw man: contradict until the opponent EXAGGERATES their own claim, then refute the exaggeration. His seven-word defense is the best anti-straw-man mantra in the course — “That’s what I said, and no more” (p. 31). Worth writing on the board.

The Ice Cream Swap (2005)

The Kid’s claim, before you watch — the only thing he asserts: chocolate is the best.

Listen for the one sentence where that claim gets swapped.

Thank You for Smoking (2005), 30 seconds, ONE move — Nick Naylor (a tobacco lobbyist) “winning” an argument against his son Joey. ORIENT (10 s): fictional, comedic; the clip opens with Nick setting the hypothetical (you defend chocolate, I defend vanilla). Read the claim line aloud before playing. The swap is one sentence at ~0:16: “So it’s ALL chocolate for you, is it?” — best → only. Play twice; the second time, stop right after the swap sentence and let them name it. AFTER the clip, quote Joey’s protest from the scene aloud: “But that’s not what we’re talking about!” — it falls just past this cut. Embedded from Dropbox; OFFLINE BACKUP: media/tyfs_icecream_short.mp4 in VLC (.en/.es subtitles auto-load); EN/ES transcripts in media/ to print. A fuller 62 s version of the scene is in media/tyfs_icecream.mp4 (adds Nick’s confession “I’m not after you, I’m after them” — optional extension for a strong group).

Nobody Said Chocolate

The move	The line
1. The claim	Kid: “Chocolate is the best.” = Among other flavors, I prefer chocolate.
2. The rebuild	Dad: “So it’s ALL chocolate for you?” = Among other foods, you would always choose chocolate
3. The attack	Dad: a speech on freedom, choice, liberty. = You are against choice and thus, liberty
4. The original claim	I prefer chocolate

One word swapped:
- “best” became “all”; that’s the whole trick.

One move, fully visible in 30 seconds: claim → copy → attack-on-the-copy. Joey’s protest (“But that’s not what we’re talking about!”) falls just after the cut — QUOTE IT ALOUD when the clip ends: it is the deck’s straw-man test performed in the scene — the opponent, asked in effect whether he accepts the summary, says no. Nick’s closing line (“Ah — but that’s what I’M talking about”) shows the tactic is DELIBERATE: he knows he changed the subject; that candor is what makes the clip safe to laugh at and easy to diagnose. If a strong group wants more: the 62 s version (media/tyfs_icecream.mp4) adds “I didn’t have to prove vanilla was best — I proved you’re wrong” (fallacy fallacy, weaponized) and “I’m not after you, I’m after them” (the audience — the closing debate). Keep the projected slide to the ONE move; the extensions are verbal.

Nobody Said Chocolate

The move	The line
1. The claim	Kid: “Chocolate is the best.”
2. The rebuild	Dad: “So it’s ALL chocolate for you?”
3. The attack	Dad: a speech on freedom, choice, liberty.
4. The original claim	Never addressed — still unanswered.

Test: would the Kid accept the following representation of the his argument?

“you’re against choice”?
No → straw man.

Red Herring

“Q: Does the curfew actually reduce crime?”
“A: What about the money this city wastes on festivals?”

The reply changes the subject — the question is never answered.
The new topic is easier to argue, so the switch feels natural.
Straw man rebuilds the claim; red herring abandons it.
Test: is the reply still about the original question?

NARRATION: The second member of the family, and straw man’s neighbor. The question is about the curfew; the reply is about festivals — no rebuilt claim, no distortion, just a new subject that quietly replaces the old one. Run the three steps: name the move — red herring; quote the words — “what about the money… on festivals”; say what would decide it — an answer about the CURFEW. The pairing to hammer: straw man rebuilds the opponent’s claim into a weaker one and attacks that; red herring abandons the claim entirely and argues something else. Both leave the original claim standing — that is why they share a family. The whatabout-reply is the most common modern form. Latin name for the genus, if asked: ignoratio elenchi — “missing the point” — the conclusion argued is not the conclusion at issue. Vaughn files red herring under irrelevant premises (Ch. 5). Fair-question guard, as with every family: raising a DIFFERENT relevant issue openly (“that is true, and there is also a budget question”) is not a fallacy — the fallacy is offering the new topic AS IF it answered the old one.

Slippery Slope

“Fail this one exam, then never graduate, and never get a job.”

One step is claimed to make the catastrophe inevitable
The earlier chain caution: every link is another premise
- ten steps, ten chances to be wrong.
Demand the middle: how exactly does A lead to Z?

“Don’t Wake Up in a Roadside Ditch” (2012)

An entire slippery slope in 30 seconds: - Count the arrows - Which one is ever argued?

Seven Arrows, Zero Arguments

ON HOLD → angry → blow off steam → accident → eye patch → “tough” → challenged → DITCH

Each single step is slightly plausible
- the chain is absurd.
The ad is honest comedy:
- at seven arrows, everyone sees it

Test: “How exactly does an eye patch lead to a ditch?”

Equivocation

P1.	Mice are mammals.
P2.	The thing next to my laptop is a mouse.
C.	The thing next to my laptop is a mammal. ✗

In P1, mouse = the animal.
In P2, mouse = the gadget.
The word changed meaning between premises

Hold up an actual computer mouse for the punchline if one is at hand — “does it have fur?” The double meaning is IDENTICAL in Spanish (ratón: animal and gadget), so the example works in both of the room’s languages — chosen for exactly that. Bonus: “mammals” places this argument back inside the morning’s Euler world; ask where on the MAMMALS diagram the gadget-mouse belongs (nowhere — that’s the point). Say aloud: a vague word alone is NOT equivocation — the fallacy is the SWITCH between premises; each premise is true under ITS meaning, and the argument pretends they used the same one. VERBAL ASIDE if the room enjoys it (30 s, keep OFF the slide): the Witch-king and Eowyn — “no man can kill me,” and a woman does. Label it PRECISELY: that is ambiguity misread, not equivocation — he interprets one ambiguous word (“man”: human vs. male) too broadly, once; no meaning SWITCHES between premises of an argument. Fun lineage: Tolkien is echoing Macbeth’s “no man of woman born,” Shakespeare’s canonical riddle. Use it as a contrast case that sharpens the criterion: two meanings existing is the FUEL; the fallacy is the mid-argument switch — the mouse, not the Witch-king.

Equivocation

The serious version — the same switch, with real stakes:

P1.	In science, evolution is a theory.
P2.	A theory is just a guess.
C.	So evolution is just a guess. ✗

In P1, theory = a tested explanatory framework.
In P2, theory = a hunch.
The word changed meaning between premises.

The test: same word? — make sure it’s the same meaning, start to finish.

“Just a theory” — the most consequential equivocation in public life, and the switch is INDISPUTABLE: in P1 “theory” is science’s word for a tested explanatory framework (gravity is one too); in P2 it is the street word for a hunch. Each premise is fine under its own meaning; the conclusion needs them to be the SAME, and they never were. Bilingual bonus: “teoría” carries both senses in Spanish. Frame it about the WORD, not anyone’s beliefs — the same trick runs on “natural,” “chemical,” “energy.” VERBAL runner-up, kept off the slide because its diagnosis is contestable: “AI undermines democratic discourse; whatever undermines democracy must be stopped at any cost; so stop AI.” A sharp student can read “undermines” consistently and simply reject P2 as disproportionate — useful as a discussion of why clean diagnosis matters. Land the mouse first for the laugh, this one for the stakes; the cure line closes both. Family: distorted claims. Vaughn files equivocation under ambiguity.

Equivocation vs. Verbal Dispute

Same starting point — one word, two meanings. Different situations:

	Equivocation	Verbal dispute
The two meanings	both inside one argument	one per side
The word is used…	inconsistently — it switches between premises	consistently — each side keeps its own meaning
What’s broken	the argument — the conclusion needed one meaning throughout	only the appearance of disagreement
The cure	same meaning, start to finish	ask: “what do you mean by ___?”

THE POSITIONING SLIDE — it pre-answers the question the room is about to have: “is this a form of equivocation?” No. Equivocation is an ARGUMENT defect: one speaker, one inference, the meaning switches between premises. A verbal dispute is a CONVERSATION defect: two positions, two meanings, each used consistently, and the arguments never touch — nobody cheats, which is exactly why the subtitle says “not a fallacy.” PRECISION if a student probes: the criterion is NOT headcount. Two people can jointly equivocate — one supplies the weak-sense premise, the other the strong-sense premise, and together they build one switching argument. The real test: is there an INFERENCE whose conclusion needs the word to keep one meaning it didn’t keep (equivocation), or two internally consistent positions whose clash is only verbal (dispute)? It is presented here because the fuel is identical (one ambiguous word) and because in the wild you must tell them apart: call a verbal dispute “equivocation” and you have accused someone of cheating who didn’t — a mini fallacy-fallacy. Escalation rule, worth saying: a verbal dispute BECOMES equivocation the moment one speaker starts exploiting both senses on purpose. WHY AMBIGUITY IS EVERYWHERE (Aristotle, Sophistical Refutations — the first fallacy catalogue, and equivocation is already on its list): we cannot bring the objects themselves into an argument; we argue with NAMES as symbols, and a limited stock of words has to cover an unlimited variety of things — so the same word inevitably carries several meanings. Consequence for the tone guard: ambiguity is a structural feature of language, NOT evidence of deception by itself. Scientific examples worth one breath each: “effective” (statistically detectable or practically important?), “safe” (zero risk or acceptably low?), “democratic” (electoral, liberal, participatory?), “AI” (a model, an interface, a company, a whole sociotechnical system?).

Talking Past Each Other

“S1: She’s successful — she earns a fortune.”
“S2: She’s not successful — she’s miserable.”

S1 means: successful = rich.
S2 means: successful = happy.
Swap the definitions in:
- S1: “She’s rich.”
- S2: “She’s unhappy.”
Likely both S1 and S2 agree

Talking Past Each Other

“S1: She’s successful — she earns a fortune.”
“S2: She’s not successful — she’s miserable.”

The fix: ask “what do you mean by success?”

The verbal dispute — the strange non-fight where both sides are right about different claims. THE TEACHING MOVE is the substitution: write both sentences with each speaker’s definition swapped in (“she’s rich” / “she’s unhappy”) and let the room see there is no contradiction — both can be true at once, so no FACT was ever in dispute. (Precision for a sharp room: they may still disagree about which standard of success MATTERS — that residue is a values question, and the closing example makes it explicit.) Relation to the family: equivocation’s twin — there the meanings switch WITHIN one argument (the fallacy); here each side keeps its own meaning consistently (a confusion, not a cheat). Not strictly a fallacy: nobody reasons badly; the arguments simply never meet. That is why it carries its own subtitle, like tautology — a neighbor of the catalogue, not a twelfth member. It BECOMES equivocation the moment one speaker exploits both senses on purpose. Hotspot words: successful, just, fair, free, natural, real, strong — abstract nouns that hide a definition. Callback: these are CONCEPTUAL claims (the E/C/N labels from early in the course) wearing an empirical costume; and once defined, many “is it just?” disputes turn out to be NORMATIVE — a values question, not a word question. The cure question is step zero of any real argument: define the term, then check whether a disagreement survives. If it does, THAT is the argument worth having.

“Is a Hot Dog a Sandwich?” — The Late Show (2017)

Two definitions, one word: - “meat between bread” vs. - “you’d cut it in half.”

Colbert vs. Hodgman, The Late Show (~44 s — cut to end on Hodgman’s definition). The purest verbal dispute on tape: no fact about hot dogs is in question — Colbert defines sandwich STRUCTURALLY (meat between bread), Hodgman by RITUAL (cut-in-half-ability is intrinsic). Run the substitution aloud: by Colbert’s definition the claim is true; by Hodgman’s it is false; both are right about their own claim. Neither budges because neither is wrong — they are answering different questions. Zero stakes is the POINT: students see the machinery nakedly here, then the next slide’s fix question transfers it to words that matter (successful, just, free — and the closing AI-cheating example, where the same split has real consequences). OFFLINE BACKUP: media/hotdog_sandwich.mp4 — VLC auto-loads the .en/.es subtitles; full uncut segment in media/hotdog_sandwich_full.mp4. Add a Dropbox raw link as a second if hosting is needed.

One Word, Two Definitions

	Colbert	Hodgman
The claim	“A hot dog IS a sandwich.”	“A hot dog is NOT a sandwich.”
His definition	sandwich = meat between bread	sandwich = a thing you’d cut in half
By his own definition…	right	right

Ask yourself: what fact about hot dogs do they disagree on?
- None
Each is right by his own definition
- there is no factual disagreement.

The breakdown, ~60 s. OPEN WITH THE QUESTION before the table: “What fact about hot dogs do they actually disagree on?” Let the room hunt — the answer is none, and discovering that IS the lesson. Then answer the natural follow-up (“so what ARE they debating?”): the definition itself — where the category’s border goes. That is a CONCEPTUAL question (E/C/N callback): no fact decides it; you decide it by convention. Harmless here because nothing rides on the border; consequential in the closing example, where laws do. Then the table: same word, two rulebooks, each right by his own. The substitution bullet repeats the move from the successful/rich slide, so the skill is now used twice. Closing line is the keeper: a fight about a word, not about the world. Bridge to the next slide: so what’s the fix? Ask what the word means — and then check whether any disagreement survives (hot dogs: none does; the closing example: one does).

The Hidden Question

“S1: Using AI to write your essay is cheating — you didn’t do the work.”
“S2: It’s not cheating — it’s a tool, like a calculator or a spell-checker.”

S1 means:
- “your own work” = you wrote every word.
S2 means:
- “your own work” = you made it — tools allowed.

The Hidden Question

“S1: Using AI to write your essay is cheating — you didn’t do the work.”
“S2: It’s not cheating — it’s a tool, like a calculator or a spell-checker.”

Swap the definitions in:
- by S1’s definition, it is cheating.
- by S2’s definition, it isn’t.

Both are consistent — so part of this fight is just the word.
- but only part.

The Hidden Question

“S1: Using AI to write your essay is cheating — you didn’t do the work.”
“S2: It’s not cheating — it’s a tool, like a calculator or a spell-checker.”

The real question:
- where is the line?
- calculator → spell-checker → Google → AI outline → full AI draft
Most people accept the calculator; most courses reject the unedited AI draft.
The real disagreement is somewhere in between — and arguing about where is a real debate, where agreement becomes possible.

The Hidden Question

“S1: Using AI to write your essay is cheating — you didn’t do the work.”
“S2: It’s not cheating — it’s a tool, like a calculator or a spell-checker.”

With charged words — cheating · tool · your own work:

Pick the definition, and the verdict follows: — “punish it” / “allow it.”
Different definitions become different normative claims.

The full lesson: defining the term removes the fake disagreement — and exposes the real one: what is school for?

The stakes-raising closer — and the one argument every student in the room has actually had, threading the course’s AI through-line. WARM-UP, 15 seconds, before the slide: “You have had this fight since age six: ‘That’s not fair — the slices aren’t equal!’ / ‘It IS fair — I’m older and hungrier!’ Nobody disputes a fact about the cake; ‘fair’ hides two definitions (equal shares vs. matched to need), and each kid’s SHOULD falls straight out of their definition. Today’s version of that fight:” — then reveal the AI quote. SYMMETRY IS THE DESIGN: both speakers are earnest, both slogans smuggle a definition, neither is the fool. Beat 1 runs the same substitution move as the “successful” slide, so the skill repeats: state each speaker’s hidden definition, swap it in, and the “disagreement” turns out to be two consistent verdicts from two definitions. END BEAT 1 DELIBERATELY INCOMPLETE — “but only part”: unlike “successful,” clearing up the word does NOT end this one. Beat 2 is the spectrum — the line does the intuitive work, and the agreement payoff is real (“brainstorming yes, final prose no”). THE FULL LESSON, and say it in these words: clarifying definitions dissolves PART of the disagreement and EXPOSES the real normative disagreement underneath — what education is for, what counts as authorship, how much assistance is permissible. The residue is a values question: what is school FOR (learning vs. credential)? NORMATIVE (E/C/N callback). Beat 3 is the arc’s pattern: charged words describe and evaluate at once, so choosing a definition silently chooses the verdict. Contrast: “successful” dissolved; sandwich was pure convention; this one becomes the real debate. Payoff for YOUR classroom: this is why the course AI policy defines permitted uses explicitly — definitions before verdicts. OPTIONAL DRILL (2 min), the compound question — Aristotle’s trick of bundling several questions and demanding one answer: put up “Why did your failed policy make crime worse?” and ask the room what must be split BEFORE answering: (1) did the policy fail? (2) did crime get worse? (3) did the policy cause it? The defense is one move: SPLIT the question, answer the parts. Same skill as the hidden question — expose what the sentence smuggles — so it slots here, not as a twelfth catalogue entry.

False Dilemma

“S1: Either we close the border — or we lose the country.”

One skeleton: [my policy] — or [catastrophe].
The catastrophe is asserted, never argued.
Only two options offered when more exist.

False Dilemma

“S1: Either we close the border — or we lose the country.”

The fix: ask “is that everything?”
- the hidden middle: visas, quotas, asylum rules, work permits…
Name a third option, and the “or” collapses.

Emotional Appeals

“We shouldn’t be so restrictive about migration — think of all those mothers separated from their children.”

Appeal to pity — the quote above: a heartbreaking case offered as sufficient to decide the whole policy question.
Appeal to fear — the reply from the other side: “Open the borders, and crime floods your streets.”
Bandwagon — “Most countries adopted lockdown measures — so, we should do the same.”

Emotional Appeals

“We shouldn’t be so restrictive about migration — think of all those mothers separated from their children.”

The tell:

in each, one feeling is the whole case
fear and pity pull in opposite directions
a feeling alone can’t decide either way.

NARRATION: The fourth family doesn’t rely on faulty reasoning, a misused source, or a distorted claim — it offers a feeling as if the feeling decided the question. Two moves on the slide, one per side. The HEADLINE quote performs the first: appeal to pity (ad misericordiam) — the mothers and their separated children. SAY THIS CAREFULLY: the suffering is real, and it is RELEVANT — consequences for rights, welfare, and dignity are legitimate considerations in any policy argument. The fallacy is not “emotion appeared”; it is one heartbreaking case offered as SUFFICIENT to decide the whole question, substituting for a proportionate look at the evidence and the competing costs. Then the reply from the other side, appeal to fear — scare tactics (NOT ad baculum: that Latin names the appeal to force, “agree or I harm you”; generic fear is broader): open the borders and crime floods your streets — dread, likewise offered as the whole case, selling restriction. Third move, bandwagon (ad populum): the inference is now EXPLICIT on the slide — “most countries do it, SO it must be right”: popularity offered as proof of correctness. That explicitness is deliberate: “many countries are tightening” could otherwise be an abbreviated evidence-based argument (they may be responding to real conditions), and only the so-it-must-be-right step makes it fallacious. Run the three steps aloud — (1) name: bandwagon; (2) guilty words: “so it must be the right policy”; (3) decide it: popularity is not outcomes — what actually happened where borders tightened? Note its special tell: unlike fear and pity it doesn’t even name a consequence — it outsources the whole judgment to the crowd. THE DEEP POINT: fear and pity here pull OPPOSITE ways — the same machinery sells either conclusion, so a feeling ALONE cannot decide the question either way. And the guard against the crude lesson, said aloud: this does NOT mean moral concern is irrational while numbers are neutral — numbers can mislead too, and compassion legitimately shapes what a policy is FOR. The test is never “was there emotion?” but “was the feeling offered as the whole case?” No side owns the fallacies: each direction got one. Vaughn files fear and pity under “appeal to emotion” and bandwagon under “appeal to popularity” (irrelevant premises). Migration case appearance #4 (labels, fair question, false dilemma, now feelings). SCHOPENHAUER ON BANDWAGON (assigned reading, Stratagem XXX, Saunders 1896, p. 37): “There is no opinion, however absurd, which men will not readily embrace as soon as they can be brought to the conviction that it is generally adopted… They would sooner die than think.” And his dissection of where “universal opinion” comes from: it “is the opinion of two or three persons,” repeated by the lazy and credulous until adhesion becomes a duty. (Caution: the longer famous version “…in fact they do so” is Russell’s riff, not Schopenhauer.) One line connects it forward: counting believers is checking the wrong gate.

Diagnosing Emotional Appeals

Demo case:

“Open the borders, and crime floods your streets.”

Name the move → appeal to fear (scare tactics).
Quote the exact words → “crime floods your streets.”
Say what would decide it → what do crime data on migrants actually show?

Run the same three-step spine used on every family, so the emotional appeals get graded exactly like the rest — naming is vocabulary, step 3 is the scientific thinking. The antidote to a manufactured feeling is the evidence it skipped: pull the actual data. The fallacy-fallacy guard matters DOUBLE on a hot topic, state it carefully: naming the scare tactic does NOT prove migration is risk-free, and naming the pity appeal does NOT prove the families’ suffering irrelevant — the labels only say the feeling was offered where evidence belongs. That hand-off is the through-line into the fallacy-fallacy slide and — the Kahneman reading explores where the fear comes from. ACCORDION: if time is short, cut this slide and run the three steps aloud on the fear line from “Emotional Appeals.”

The Gish Gallop

Not one broken argument — too many to check.

Bury the point under a flood of weak claims, faster than anyone can answer.
Every claim left unanswered reads as a point scored.
A tactic, not a fallacy — it can carry any move on the map, or none.

The full anatomy — the criteria for calling one, and the defense — comes in the first session after the midterm.

NARRATION: The closing meta-tactic — not a bad inference but the sheer NUMBER of them. Volume works because refuting costs far more than asserting (Brandolini’s adage — an aphorism, not a measurement): in a timed exchange, every unanswered claim reads as a point scored. Named for creationist Duane Gish, who used it in debates. GUARD, one sentence: many claims alone is NOT a gallop — calling one takes criteria, and those criteria, the debate clip, the psychology, and the defense all belong to the first session after the midterm; today, only the name and the caution. Keep Gish OUT of the family list — it is the volume knob, not a family of its own. If a student remembers the claim-wall from the earlier debate session, congratulate them: that wall gets named and dissected then. The drill does NOT test the gallop — bonus, not exam.

Drill: Name That Fallacy

“I ate cereal before the exam and got an A. Cereal before exams gives you A’s.”
“Don’t trust Luis’s math solution — he failed English.”
“I think we should eat less meat.” — “So you want everyone to be vegan? Ridiculous.”
“Stars produce their own light. Shakira is a star. So Shakira glows in the dark.”
“Either you support the curfew, or you don’t care about safety.”
“Professor, I studied all night — you can’t fail me.”
“He’s the best player ever — most goals in history.” — “No: he never won a title.”
“The bridge inspector found deep cracks. She warns: don’t drive across.”

NO LABELS ON THIS SLIDE — that is the drill; the answer bank is the NEXT slide, flip to it while pairs work. Steps 1–2 of the three-step diagnosis on all eight (fast, called out); then pick ONE and run step 3 (what would decide it). STRUCTURE (for you): 2 recap from the previous session, 4 on today’s families, 1 verbal-dispute case, 1 NOT-a-fallacy contrast — the last two are the point. ANSWER KEY: - 1. FALSE CAUSE (recap) — “after” is not “because”; decide it: grades with and without cereal. - 2. AD HOMINEM (recap) — English grades are irrelevant to the math; the person is attacked, the solution never checked. - 3. STRAW MAN — “less meat” rebuilt as “everyone vegan,” then mocked; would the speaker accept that summary? No. - 4. EQUIVOCATION — “star”: celebrity vs. celestial body; the word switches meaning between premises (“estrella” carries both in Spanish too). - 5. FALSE DILEMMA — supporting this curfew and caring about safety are not the only two options; name a third: other safety policies, or opposing it on the evidence. - 6. APPEAL TO PITY — the effort is real and may deserve sympathy, but it is not evidence the answers were right; one feeling offered as the whole case. - 7. VERBAL DISPUTE — the trap: it is NOT on the answer bank. “Best” = most goals for S1, most titles for S2; swap the definitions in and both are right. Partly a word fight — and partly the real question: what SHOULD “best” weigh? - 8. NO FALLACY — the second trap. A documented risk, an expert in exactly her field, evidence you could check: legitimate fear, legitimate authority. If someone cried “scare tactics!” here, THAT would be the error (fallacy-fallacy territory). This is the item that separates diagnosticians from label machines. Multi-label answers are FINE if defended. Debrief closer, verbatim: “You named the moves. Did you thereby prove any of these claims false?” (No — fallacy fallacy, next.)

The Answer Bank

Your options:

Faulty reasoning	false cause · circular reasoning
Misused sources	ad hominem · appeal to authority
Distorted claims	straw man · red herring · slippery slope · equivocation · false dilemma
Emotional appeals	appeal to fear · appeal to pity · bandwagon

Careful: one item isn’t on this menu — and one isn’t a fallacy at all.

The Same-Claim Test

A genuine refutation must contradict the same claim (Aristotle, Sophistical Refutations):

Same…	Ask
Meaning	are the key words used in the same sense?
Subject	the same person, group, population?
Outcome	the same result being measured?
Comparison	against the same baseline?
Time	during the same period?

Claim: “The policy reduced violent crime in 2025.” — Reply: “But burglaries rose in one district in March.” Both can be true; nothing was refuted.

The speaker changes one coordinate — and hopes nobody notices.

The unifying tool for the whole catalogue — and the bridge into the fallacy fallacy next: both slides police the same urge, declaring victory before the work is done. To refute a claim you must contradict THAT claim — same meaning, subject, respect, comparison, time. Change one coordinate and you get the LOOK of a refutation with none of the substance. Roll-call, aloud: straw man changes the CLAIM; equivocation changes the MEANING; cherry-picking changes the population or the TIMEFRAME; a dodge or a joke changes the TASK entirely (a future session dissects the most famous one on tape). The worked example: both statements can be true — the reply switches crime category, unit, and window, so it never touches the claim; have the room name which coordinates moved (subject, time). Schopenhauer’s Stratagems I, III, and XXIII (assigned reading) are all same-claim violations, and his defense line is this test in seven words: “That’s what I said, and no more” (p. 31). The example is deliberately abstract policy — no party attached.

The Fallacy Fallacy

“Your argument for the curfew is a slippery slope. So the curfew is useless.” ✗

Showing an argument invalid does not show its conclusion false.
A true conclusion can come from a bad argument.
Labels are diagnostic tools, not refutations.
Real arguments may fit several labels — or none cleanly.

The label only starts the work.
What decides it: the evidence

THE guard against label-machine students — teach it with the same weight as the catalog. Two-step habit: (1) name the broken move; (2) ask what evidence WOULD decide the conclusion. The closing debate grades both steps. This also legitimizes multi-label answers in the drill: the point is the diagnosis, not the taxonomy trivia. SCHOLARLY BACKING if the room wants authority (Mercier & Sperber 2017, pp. 228–231): “most if not all fallacies on the list are fallacious except when they are not” — context decides. Their twin pair to borrow: “you drink too much, so your warning about my drinking is worthless” is a tu quoque that FAILS (the drunk’s warning can be sound); but “you say chocolates are sinful yet you eat them” fairly challenges the speaker’s own commitment. Same label, opposite verdicts — experimental subjects track the evidential strength, not the label (Hahn & Oaksford, cited there). The label starts the work; it never finishes it. AND THE MOVE IS IN THE ASSIGNED READING TWICE (Schopenhauer, Saunders 1896): Stratagem XXXVII — refute the opponent’s WORST proof and claim the whole position falls (“bad advocates lose a good case,” p. 45); and Stratagem XXII — cry “begging the question!” at a legitimate premise to block it (p. 30). The fallacy fallacy is old enough to be in the cynic’s manual — as a recommended attack.

Conclusion

Name the move, quote the words, say what would decide it.

Distorted claims — straw man, red herring, slippery slope, equivocation, false dilemma.
Talking past each other — one word, two definitions; define the term first.
Emotional appeals — fear, pity, bandwagon: a feeling where the evidence should be.
The Gish gallop — too many claims to check: a tactic, analyzed after the midterm.
The fallacy fallacy — naming the move never proves the conclusion false.

The one-slide recap of TODAY: the distort family (four moves), the verbal-dispute neighbor (the definitional non-fight — hot dogs to AI essays), the feelings family (three moves, aimed both ways on migration), the volume meta-tactic with its three documented findings, and the fallacy-fallacy guard. Say the unit tally verbally: four families across two sessions — fake, play, distort, feelings — plus the twins from the formal session. Tone guard, said once: “cheat” is the frame, but most fallacies are committed sincerely — careless or motivated reasoning, not conscious deception. The thesis line IS the spine; the fallacy-fallacy bullet is the trap. Close on the teaser but DO NOT resolve it — “the seller is your own mind” is the Kahneman reading’s territory. Announce homework verbally: the exercise set, parts (a)–(b); read Kahneman Ch. 1.

References

Aristotle. On Sophistical Refutations. (any scholarly edition)