Logic and Scientific Thinking

Question	Answer from the previous session
Valid form vs. true conclusion?	A valid shape can still carry a false claim
The formal twins?	Four traps, each mimicking a valid shape (+ one lone tool)
The bat and the ball?	The felt answer ($0.10) wasn’t the checked answer ($0.05)

Formal vs. Informal

	Where the failure shows
Formal fallacy	in the argument’s form alone — no content needed
Informal fallacy	only in the content: relevance, ambiguity, evidence, context

A fallacious argument can still have a true conclusion — and can even be formally valid.

Most fallacies aren’t lies: careless or motivated reasoning, not conscious deception.

NARRATION: One bridge back before the catalogue. The earlier verdict machine grades the argument; today names HOW it fails. But do NOT map verdicts to fallacy types one-to-one (invalid=formal, unsound=informal is too neat and wrong). The honest line: a FORMAL fallacy is visible in the shape alone — write it in letters and the error shows (the four twins). An INFORMAL fallacy only shows when you read the content: relevance, ambiguity, evidence, context. Killer example, coming shortly: circular reasoning. “P, therefore P” is formally VALID — yet worthless, because it gives no independent reason. That is exactly why it is filed INFORMAL. Fragment 2 seeds the fallacy fallacy: fallacious does not mean false conclusion, and does not even mean invalid. Fragment 3 is the tone guard for the whole catalogue: “fighting dirty” is the fun frame, but most fallacies are committed sincerely — careless or motivated reasoning, not conscious deception; scholars do it too.

The Three-Step Diagnosis

Demo case:

“Crime fell after the curfew. The curfew works.”

Name the move → false cause.
Quote the exact words → “fell after” becomes “works.”
Say what would decide it → could there be other causes behind crime falling?

The Map: Four Families

Twelve informal tricks, colour-keyed by family — plus one 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

The full line-up, colour-keyed to the four families — the visual taxonomy for the session. The four formal twins came previously; today’s twelve informal moves make up Act II’s field guide. Say the caveat aloud once: our four families are a practical filing system, not canonical — “faulty reasoning” means the SUPPORT fails, not “invalid form”: neither false cause nor circular reasoning is a structural error (circular reasoning is even formally valid). A sharp student may object to the cages; agree, and point to the caption. The Gish Gallop sits apart in grey — not a family member but the volume meta-tactic from the earlier “Flood,” carrying any move above; it returns as a later session’s endnote. Vaughn (textbook backup) sorts differently: Ch. 5 “irrelevant” vs. “unacceptable” premises, files the formal fallacies in Ch. 6 (p. 202) not Ch. 5, calls equivocation the “fallacy of ambiguity,” and splits ad hominem into four subtypes (p. 155). Same animals, different cages; the exam uses OUR four families.

The Map: Four Families

Twelve informal tricks, colour-keyed by family — plus one 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 — returns in a future sessionmeta

False Cause

“Crime fell after the curfew. The curfew works.”

Post hoc ergo propter hoc:
- after this, therefore because of this.
Recall: the vitamin C “cure.”
Possible rivals:
- prior trends
- coincidence
- simultaneous policies
- selection

“Not a Bear in Sight” — False Cause

Run the three steps on Homer
Lisa’s rock is the counterexample method.

The Simpsons, “Much Apu About Nothing” (1996), 27 seconds — play twice if needed; the whole argument is in the first line. Embedded from Dropbox; OFFLINE BACKUP: media/simpsons_bear_patrol.mp4 in VLC (loads .en/.es subtitles automatically). EN/ES transcripts in media/ to print for L2 students. ANSWER KEY: (1) Name — false cause (post hoc): no bears after the patrol, so the patrol works. (2) Guilty words — “not a bear in sight… must be working like a charm.” (3) What would decide it — bear sightings BEFORE the patrol; comparable neighborhoods without one. Lisa skips to the counterexample: the tiger rock has identical logic and an absurd conclusion — one imagined world where the premises hold and the conclusion fails. Homer buying the rock = the fallacy surviving its own refutation; savor it.

The Rock Proves Nothing

	Homer’s patrol	Lisa’s rock
Claim	The patrol keeps bears away.	The rock keeps tigers away.
Evidence	No bears in sight.	No tigers in sight.
Logic	It’s here, and they’re not → it works.	Exactly the same.

“Nothing happened after X” does not show X prevented it.
There were never bears around
- the patrol gets credit for nothing.
Lisa’s rock: same argument, absurd conclusion → the argument is broken.

The Rock Proves Nothing

	Homer’s patrol	Lisa’s rock
Claim	The patrol keeps bears away.	The rock keeps tigers away.
Evidence	No bears in sight.	No tigers in sight.
Logic	It’s here, and they’re not → it works.	Exactly the same.

That is false cause: “no bears after” does not, by itself, show the patrol keeps them away.

Circular Reasoning

“Smoking is unhealthy — because it damages your health.”

The reason gives no independent support
- to accept it, you must already accept the claim.
Here the reason just restates the claim

The test: does the reason stand on its own without already believing the conclusion?

NARRATION: Circular reasoning — its Latin name is petitio principii, “begging the question.” The argument offers a reason, but the reason is just the conclusion wearing different clothes. “Smoking is unhealthy because it damages your health” — unhealthy and damages-your-health are the same claim; nothing new was added, so the circle never touches the ground. The test: does the reason stand on its own — something you could verify WITHOUT already accepting the conclusion? And notice the conclusion here is TRUE. A circular argument can carry a true conclusion on a broken vehicle — which is exactly why naming the move is not the same as refuting it (the fallacy fallacy, next session). Family: filed under “faulty reasoning” because it only POSES as support — but be precise (sharp students will be): “P therefore P” is formally VALID. The failure is not the shape; it is epistemic — zero independent support. That is exactly why it is an INFORMAL fallacy. General principle: a circular argument’s support eventually depends on already accepting its conclusion. Restatement (this slide) is only the most obvious version — the cover-up slide shows a circle through two DIFFERENT claims, no restatement at all. Everyday-usage flag: “begs the question” is now widely misused to mean “raises the question”; here it means assuming the point. Sharp-student callback (keep OFF the slide): the earlier Romney premise “fewer ships means weaker” only begs the question if you DEFINE capability as ship-count — separate the two and it stops being circular.

“It’s Got Electrolytes” — Idiocracy (2006)

Run the three steps:
notice the reason and the claim are the same sentence.

Idiocracy (2006), the Brawndo/electrolytes scene — ~67 s. The perfect circular loop: Brawndo is good for plants BECAUSE it has electrolytes; electrolytes are what plants crave BECAUSE that is what is in Brawndo. The reason never leaves the claim. ANSWER KEY: (1) Name — circular reasoning (begging the question). (2) Guilty words — “it’s got electrolytes… it’s what plants crave… it’s what they use to make Brawndo.” (3) What would decide it — an INDEPENDENT test: do plants actually grow better with Brawndo than with water? (In the film they don’t — the crops are dying.) Tie back to the smoking slide: the reason has to stand on its own. OFFLINE BACKUP: media/idiocracy_brawndo.mp4 — VLC auto-loads the .en/.es subtitles; EN/ES transcripts in media/ to print for L2 students. The &raw=1 Dropbox link streams the hosted copy.

What Plants Crave

	Brawndo’s pitch (the film)	A real reason
Why use it?	“It’s got electrolytes.”	“Brawndo plots out-yield water.”
Where’s the reason from?	“electrolytes” = “what plants crave” = the claim	an independent test
The move	circular reasoning	a checkable argument

Four turns, zero new facts: each “reason” just renames the claim.

Test: does the reason give you anything you could check without already accepting the claim?

Right after the clip, ~60 s. Build the left column first: walk the film’s answer — “electrolytes… what plants crave… it’s got electrolytes” — and ask the room what NEW fact ever arrived. None; every turn renames “it’s good for plants.” Then the right column: a real reason comes from OUTSIDE the claim — a field trial you could run and lose. The test is INDEPENDENCE: a genuine reason is checkable without already accepting the claim — a field trial could vindicate Brawndo or sink it. (For pure restatement circles a sharper test also works — the “reason” cannot even be false while the claim is true, because it IS the claim — but do not teach that as the general test: longer circles, like the cover-up next, pass it.) Tie forward to “The Circle Never Lands,” which contrasts straight vs. circular on smoking.

“It’s a Cover-Up” — Circular Reasoning, in the Wild

“There’s no proof because it’s all covered up —
and we know it’s covered up because there’s no proof.”

No restatement here: two different claims, each defended by the other.
Self-sealing: nothing could ever count against it.

NARRATION: Circular reasoning at its most seductive — the self-sealing conspiracy. “No proof because it’s covered up; we know it’s covered up because there’s no proof.” Demand evidence and the missing evidence gets folded back in as confirmation. This is the important variant: neither sentence restates the other and neither is a tautology — the circle runs through the SUPPORT, not the wording. Use it to kill the idea that circularity = synonyms. A belief that explains away every possible objection has stopped saying anything about the world. This threads the course’s AI-misinformation spine: the conspiracies that feel most airtight are often the emptiest, precisely because they immunize themselves against disproof. The cure is on the next slide — does any reason stand on its own, checkable without already believing the conclusion? REAL CASE, one breath (Mercier & Sperber 2017, on the Dreyfus affair): Bertillon’s “auto-forgery” theory — if Dreyfus’s handwriting matches the incriminating document, he’s guilty; if it doesn’t match, he disguised it — so he’s guilty. Same self-seal, convicting an innocent man: every possible observation confirms, so the theory says nothing.

Straight vs. Circular

Straight — the reason stands on its own:

Smoking is unhealthy — cohort studies show.
Checkable.

Circular — the reason just restates it:

Smoking is unhealthy — because it damages your health.

Tautology vs. circular reasoning — a tautology is a vacuous statement (“it is what it is”); circular reasoning is an argument whose support loops back on its conclusion. Some circles lean on a tautology — but the concepts are distinct.

True, but Empty

“Either it rains tomorrow, or it doesn’t.”

Always true — there’s no way for it to be wrong.
And it tells you nothing
That’s a tautology: true because it says nothing about the world.

Platitudes — the everyday versions:

“Boys will be boys.”
“A rule is a rule.”

The clean tautology, right after the distinction. Weather line first: “Is it true?” — yes, guaranteed. “Does it help you plan tomorrow?” — no. It is true BECAUSE it rules nothing out. Then the everyday slogans, named PLATITUDES: trite phrases that sound wise because they cannot lose. Nuance if a student pushes: some are empty by FORM (tautologies proper — “it is what it is”) and some by VAGUENESS (“everything happens for a reason,” “we need common-sense solutions”) — not strictly tautological, just unpinnable. And “boys will be boys” is sneakier still: in real use it smuggles a NORMATIVE claim — that the behaviour should be excused — a loaded truism, not a pure logical truth (callback to the earlier empirical/conceptual/normative labels). Same diagnostic for both: what would it rule out? Nothing → it said nothing. Tie back: a circular argument’s “reason” IS a tautology dressed as evidence — that is why it proves nothing. Political example, verbal: “Brexit means Brexit” — X means X, firm-sounding while committing to nothing (mocked left and right, so no side owns it). The tell — nothing to attack, because nothing was said.

Ad Hominem

“The mayor’s numbers show the curfew cut crime; The guy’s a crook. Ignore it.”

Only the arguer is hit
- the numbers go unanswered.
Even if “he’s a crook” is true, it doesn’t make the numbers wrong.

Running case, clearest form: a real argument is on the table (the crime numbers), and the reply just attacks the man and walks off — the tell is “ignore it,” the evidence never opened. Sharpen the point: the insult might even be TRUE, and the numbers still stand or fall on their own. The clip next slide is an earlier-session callback (pure abuse); then the legitimate/fallacious table. Gradient to draw verbally: pure abuse with no argument at all (Barnard) → a jab riding an argument (Obama) → an argument dismissed by hitting the arguer (the mayor). The drill’s Luis item (“failed English”) tests exactly this line. SCHOPENHAUER’S SHARPER CUT (assigned reading, Saunders 1896, pp. 45–46): what the course calls abusive ad hominem he names argumentum ad PERSONAM — “passing from the subject of dispute, as from a lost game, to the disputant himself” — and distinguishes it from his ad hominem proper (Stratagem XVI): arguing from the opponent’s OWN admissions and conduct, which is legitimate pressure but “valid only for this particular man” — it never establishes the claim’s truth. His defense against the personal attack (p. 47): “That has no bearing on the point in dispute” — then return to it.

“This Is Abuse” — Named at Last

The mystery room you met previously finally has a name: attack the person, and the argument never happens.

Monty Python, “Argument Clinic” (1972), the abuse room — 36 s, opens on Barnard’s “What do you want?” (students saw the full sketch previously). THE CALLBACK the syllabus planned: back then this room had no name; today it is the purest ad hominem — the person is attacked and no argument is ever engaged, because none exists. The clinic even files it as a SEPARATE ROOM from argument — the sketch knows the difference; so should we. Gradient to draw verbally, pointing at the previous slide: pure abuse (Barnard) → jab riding an argument (Obama) → dismissal via motives (the mayor). Only the last two touch an argument; all three play the person. Embedded from Dropbox; OFFLINE BACKUP: media/argument_clinic_abuse.mp4 in VLC (.en/.es subtitles auto-load); EN/ES transcripts in media/ to print.

Different Rooms

	Barnard’s room	“Ignore the mayor’s numbers”
Attacks the person?	Yes — “you vacuous, toffee-nosed pervert”	Yes — his motives
Is there an argument?	No — none was ever offered	Yes — the crime numbers
The move	Pure abuse	Ad hominem — the attack dismisses the argument

The clinic files abuse and argument in separate rooms.
The fallacy begins when the attack replaces the answer to a real argument.

Test: was there an argument — and did attacking the person make it disappear?

90 seconds, right after the laugh — the sketch already made the point: even the clinic knows abuse and argument belong in different rooms (“You want 12A, next door”). Build the table left column first, then ask: “So was Barnard committing a fallacy?” No — nothing was argued, so nothing was dodged. Rude, not fallacious. Then the right column: the mayor quote has a real argument (the crime numbers), and the attack is used to make it vanish — THAT is ad hominem. The middle case, verbally: Obama’s jab rode ON an argument he also engaged — insult plus argument is neither pure abuse nor a fallacy; it’s rhetoric. This slide is the charity guard: students who call every insult “ad hominem” are being label machines — the fallacy needs a victim, and the victim is an argument.

Labels, Not Answers

“We should restrict migration.” — “That’s racist!”
“We should expand welfare.” — “That’s communism!”

The label replaces the answer.
“Woke!” · “Fascist!” · “Denier!” · “Fake news!”
The label might even fit — the argument still goes unanswered.
The sign: the conversation ends once the label is applied.

When It’s Fair

Speaker 1 claims…	Speaker 2 answers with…		Fallacious dismissal
Speaker 1 claims…	Fair question	Name-calling	Fallacious dismissal
“This study shows the curfew works.”	“Who funded the study?”	“Corporate nonsense!”	“It was bank-funded, so the result is false.”
“This expert backs the curfew.”	“Is crime actually his field?”	“That’s BS!”	“He’s not one of us, so ignore him.”
“Restrict migration — here are my numbers.”	“Would it discriminate in effect?”	“That’s racist!”	“It’s discriminatory, so ignore his numbers.”
	Weighs the testimony	No argument — just a label	Dodges the analysis

Rule: questioning a source can reduce the weight of testimony

It does not, by itself, show the conclusion false.

To answer the conclusion:

engage the reasons
the evidence
inferential link.

FRAGMENT 1 is the pairing insight: in every row the name-calling is the rude twin of the fair question — the same underlying concern (funding, competence, discrimination) with the inquiry stripped out. The test between them: could the answer change your mind? The fair question can lose (“independently funded, pre-registered” — trust goes UP); the label cannot. Corollary said aloud: the same WORDS can commit the fallacy — “who funded THIS, huh?” asked rhetorically is a dismissal in the fair question’s clothes. The precision slide for the whole “misused sources” family. Each row now opens with what Speaker 1 actually says, then Speaker 2’s three possible replies — read one row aloud as a mini-dialogue before generalizing. Bias and incentives are legitimate INFORMATION about testimony (the earlier credibility toolkit); the fallacy begins at dismissal-without-engagement. This table also pre-answers the smart-student objection “but funding DOES matter” — yes: left column. The third column completes the gradient from Different Rooms and Labels, Not Answers: name-calling engages nothing — like Barnard, it is rude rather than fallacious when no argument is on the table, and it becomes the label-dismissal fallacy the moment it is offered AS the answer to one. Left to right: weigh, abuse, dodge — the extremes first (fair vs. bare label, easy to tell apart), the subtle dismissal last, because it is the one that LOOKS like reasoning and is hardest to spot. Row 3 carries the migration case across the spectrum — same anatomy as the previous slide’s opening quote: the effects question WITH evidence is legitimate; the label as a full stop is the fallacy.

Appeal to Authority

“The police chief says the curfew works. Case closed.”

The fallacy: status substitutes for evidence
Another example: a Nobel physicist pronouncing on medicine (wrong field)
Legitimate: citing expert consensus, in its field, with the evidence available.
Watch for vague appeals to authority: “studies show,” “experts agree”
- which studies?
- which experts?

“What Cigarette Do You Smoke, Doctor?” (1949)

A real ad, run for a decade. What is the doctor’s coat actually doing in this argument?

The Lab Coat

	What the ad shows	What it proves
The authority	Doctors — “all branches of medicine”	Right field for health claims…
The claim tested	Which brand doctors smoke	…but no health claim is ever made
The survey	“Repeated, nationwide” — real	Doctors got free cartons of Camels, then were asked their brand

The doctor’s coat adds trust; the ad never gives evidence.
A real survey can still measure nothing that matters.
The unstated conclusion (safe to smoke) wasn’t tested.
Test: is the authority giving evidence?

The survey scam, documented (Stanford tobacco archive; Center for the Study of Tobacco and Society): R. J. Reynolds reps handed out free cartons of Camels at medical conventions, then surveyed doctors on their brand — some asked which pack was in their pocket at that moment. 113,597 physicians, methodology never described in the ads. The deep lesson for step 3: the survey was REAL and the reported result plausibly TRUE — and it still supports nothing, because brand preference is not a safety judgment. Authority + true fact + wrong question = the polished form of the fallacy. Tie back to the slide before: “status substitutes for evidence” — here the status is a costume literally rented by the advertiser. Future-session preview: what WOULD make expert testimony evidence (consensus, in-field, methods shown). WHY THE COAT WORKS, the lab result (McGuire, quoted in Zaller 1992, p. 45): the same message is “judged as fairer, more factual, more thoroughly documented, its conclusion following more validly from its premises, and even more grammatical” when ascribed to a high-credibility source. Source cues corrupt the LOGICAL judgments themselves — “even more grammatical” gets the laugh and makes the point: the coat doesn’t just add trust, it makes the argument LOOK more valid than it is. THE 1830s VERSION, from the assigned Schopenhauer (Stratagem XXX, Saunders 1896, p. 37): the French curé who, to dodge paving the street in front of his house like every other citizen, “quoted a saying which he described as biblical: paveant illi, ego non pavebo. That was quite enough for the municipal officers.” A FABRICATED authority (the verse doesn’t exist; the Latin is a pun — “let THEM tremble/pave”) — and it worked because, as Schopenhauer notes, the opponent “has no books at hand.” The Camel survey is the industrial descendant.

Conclusion

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

Faulty reasoning
- false cause (after ≠ because)
- circular reasoning (no independent support — the loop returns to the claim)
  - often via tautologies and platitudes: true because they say nothing; they prove nothing.
Misused sources
- ad hominem
- appeal to authority

Logic and Scientific Thinking

Welcome

Learning Outcomes

Overview

Recap

Act II: The Informal Fallacies

Formal vs. Informal

The Three-Step Diagnosis

The Map: Four Families

The Map: Four Families

False Cause

“Not a Bear in Sight” — False Cause

The Rock Proves Nothing

The Rock Proves Nothing

Circular Reasoning

“It’s Got Electrolytes” — Idiocracy (2006)

What Plants Crave

“It’s a Cover-Up” — Circular Reasoning, in the Wild

Straight vs. Circular

True, but Empty

Ad Hominem

“This Is Abuse” — Named at Last

Different Rooms

Labels, Not Answers

When It’s Fair

Appeal to Authority

“What Cigarette Do You Smoke, Doctor?” (1949)

The Lab Coat

Conclusion