Logic and Scientific Thinking

Persuasion and Rhetoric

Bogdan G. Popescu

Tecnologico de Monterrey

Welcome

The persuasion session, built on one question: HOW does a message move a mind? Arc: the puzzle (Reagan 1984 — a reply that persuades without answering) → the process (Zaller: receive, accept, sample) → the three appeals as mechanisms (ethos, pathos, logos — each with its conceptual job, its honest form, and its counterfeit) → the appeals interlocking, retraced on the joke → whose hands (misinformation / disinformation / bullshit as the sender’s orientation — same machine, different senders) → the defense (three questions; the social defense) → the dissection exercise. TIMING (~65–70 min): if pressed, fold Interlocking into the Retrace and run the exercise whole-class. The previous session’s hand-off (“who floods on purpose?”) is answered in Whose Hands — flag that at the recap so nobody thinks it was dropped.

Learning Outcomes

Trace how a message moves a mind: attention → trust → feeling → conclusion.

Name the appeals’ honest and counterfeit forms — ethos · pathos · logos.

Sort the senders behind the same machine — and apply the three questions.

Recap

Question	Answer from the previous session
Misinformation?	False premises — the wrong number, not the wrong link
The gallop?	Volume that hides false premises and missing links
The defense?	One claim, its evidence — and inoculation

Previously we asked: who floods on purpose? First, a deeper puzzle: how does a message persuade at all?

The Puzzle

“Youth and Inexperience” — Reagan vs. Mondale (1984)

One sentence, zero argument — and watch the other podium.

THE OPENER — no theory has been taught yet; that is the point. ORIENT, 15 seconds, before playing: October 1984, Reagan is 73 — the oldest president in history — and in the previous debate he seemed lost and tired. A journalist asks the question directly. Then play (~39 s): Trewhitt’s full question — stamina in a nuclear crisis, grounded in the first debate — then the punchline. WATCH FOR: the split shot — Mondale, the opponent the joke defeats, laughing along. Mondale later said he felt the race slip away in that moment — his own hindsight account, not a measurement; two weeks later Reagan won 49 states (an election with many causes — don’t let the joke absorb all the credit). This is the payoff of an old promise: the earlier debate session said “when checking is hard, what decides is confidence (and jokes)” — this is the joke. ZALLER’S OWN 1984 CASE (pp. 297–99): the dominant message that year was the crafted image — media advisors “project the idea that Reagan is the commanding leader” — while the countervailing “uninformed and uninterested” message, being fainter, reached only the most politically aware. Result: MODERATELY aware Republicans were the most likely to rate Reagan “extremely knowledgeable”; the most aware didn’t buy it. The clip is the image machine at work — and the three-filter model (coming in The Process) predicts exactly who it worked on. OFFLINE BACKUP: media/reagan_age_joke.mp4 (VLC auto-loads .en/.es subtitles); the &raw=1 Dropbox link is the hosted fallback. Uncut source in media/reagan_age_joke_source.mp4 (hosted: dropbox.com/scl/fi/3j40xqir0fi0dy2mgelyn/ …rlkey=z7y2hjer2ge12ixt09hd0u9zu).

The Puzzle

The question	“You are already the oldest president in history… Kennedy went days without sleep in the Cuban Missile Crisis. Could you function?”
The answer (Reagan)	“I will not make age an issue… I am not going to exploit my opponent’s youth and inexperience.”
The result	The hall laughs — opponent laughs — the question isn’t pressed again that night.

The question was legitimate;
The reply contains no argument
The reply left no opening for the moderator or the opponent to press on

Today’s question: how can a reply persuade without answering?

Freeze the specimen on the table before any theory. Row 1 matters most: the question was FAIR (Trewhitt built it carefully — age, the first debate, the Cuban Missile Crisis standard), so this is not a gotcha dodged; it is a legitimate inquiry dissolved by charm. Rewind the clip if the room resists row 2 — there is genuinely nothing checkable in the answer. CHARM-CAPTURE, and use it instead of fighting it: students will side with Reagan and remember the wit — point out that the room just got persuaded by a reply with zero content, which is exactly the puzzle. DO NOT resolve anything yet: the whole session builds the answer, and the joke returns, retraced, once the tools exist. Keep the claim modest: what the clip documents is that night’s exchange; the election had many causes. No later-Alzheimer’s “vindication” angle — contested for the in-office period, and it drags the room into exactly the dispute the course avoids. Symmetry note, say it once: the flooding clip (previous session) showed a Democrat, this shows a Republican — no side owns the tactics.

The Process

The Three Filters

Step one of the answer: a claim must pass three filters inside each listener before it becomes opinion:

%%{init: {"theme":"neutral","themeVariables":{"fontSize":"16px"},"flowchart":{"useMaxWidth":true,"nodeSpacing":45,"rankSpacing":50}}}%%
flowchart LR
  M["a claim arrives<br/>(news, ad,<br/>rumor, post)"] --> F1["FILTER 1: ATTENTION<br/>does it reach<br/>you at all?"]
  F1 --> F2["FILTER 2: IMPRESSION<br/>how does it land?<br/>fast, automatic"]
  F2 --> F3["FILTER 3: FIT<br/>your beliefs,<br/>your group"]
  F3 --> P["added to your<br/>stored ideas"]
  P -- when asked --> O["your answer:<br/>ideas at hand"]
  A1["most claims stop here<br/>topic interest<br/>raises the odds"] -.-> F1
  A2["confident sounds right<br/>repeated feels true<br/>feeling comes first"] -.-> F2
  A3["agreeable claim:<br/>can I believe it?<br/>disagreeable claim:<br/>must I believe it?"] -.-> F3
  style A1 fill:#f3ded9,stroke:#b44527,color:#b44527
  style A2 fill:#f3ded9,stroke:#b44527,color:#b44527
  style A3 fill:#f3ded9,stroke:#b44527,color:#b44527
  style O fill:#cfe0d8,stroke:#4a7c6f

Mechanisms: fast-vs-slow thinking (Kahneman 2011) · shortcut persuasion (Petty & Cacioppo 1984) · motivated reasoning (Kunda 1990; Gilovich 1991). Storage, sampling, and topic awareness: Zaller (1992) — his terms: reception, acceptance, “considerations,” “sampling.”

SLIDE-TO-ZALLER MAPPING, say it once: Filter 1 is his RECEPTION; Filters 2 and 3 unpack his ACCEPTANCE into what psychology distinguishes — the fast first impression (Kahneman’s System 1; Petty & Cacioppo’s peripheral cues) and the predisposition match (motivated reasoning: Kunda; Gilovich’s “can I believe it?” vs. “must I believe it?”). Storage and sampling are Zaller’s, unchanged. The rest of this note keeps his gate vocabulary. Zaller’s own model, in plain dress: RECEIVE (exposure plus comprehension — awareness RAISES THE PROBABILITY of reception; nothing switches on or off) → ACCEPT (a message is resisted only when predispositions — values, ideology, group loyalties, broader than party — plus enough contextual cues let the receiver RECOGNIZE the clash; otherwise it slips in) → accepted content is stored as a CONSIDERATION → when asked, people SAMPLE and blend the considerations most ACCESSIBLE at that moment; recency raises accessibility, but it is not the whole rule. The governor boxes say this in student language: following politics raises the odds a message gets in; beliefs plus warning cues power the resistance. BRIDGE FORWARD, one sentence: the filters say WHERE persuasion happens — Aristotle, next, says WHAT pushes a message through them. DEFINE “POLITICAL AWARENESS” when the node appears — students will ask: a person’s habitual cognitive engagement with politics — how much political communication they routinely receive AND actually comprehend. A stable trait of the receiver, not per-message attention; Zaller measures it with simple factual-knowledge quizzes (name your senator, which party holds the House) — deliberately NOT education, NOT self-reported interest — because knowledge tests capture what stuck. Distribution: “high variance around a generally low mean” — 24% of his respondents couldn’t identify the sitting vice president. That is why Gate 1 is mostly closed for most people, flood or no flood. And it governs BOTH gates in opposite directions: awareness raises reception and supplies the cues to resist — the inverted-U again. THE CLOSING FRAGMENT anchors the model in the specimen: the joke was RECEIVED by everyone in the hall (attention was already won); ACCEPTED with the laugh (no warning cue fired — Mondale’s own laugh disarmed the room’s one natural cue); and then SAMPLED — whenever the age issue threatened to resurface, what came to mind first was the joke, not the worry. That is why the question wasn’t pressed again: the accessible consideration was the laugh. THE CATALOGUE MAPPING, verbal if wanted, is OURS, not Zaller’s — he never discusses flooding. The license: his relative-intensity principle (p. 190) — attitudes track the RELATIVE intensity of competing message streams — so a flood RAISES THE ODDS that the louder stream is the one received, by drowning the counter-stream, not by jamming the receiver (the jamming story is the previous session’s load psychology). And the ad hominem is, in his own vocabulary, a CUEING message — contextual information about a message’s implications — it supplies a cue that helps receivers recognize and reject; it works only when the receiver can connect the cue to a predisposition. Fear and pity are gate-2 tools too (feelings as acceptance-shortcuts), while repetition quietly greases both gates (that last claim is the illusory-truth literature from the previous session — NOT Zaller’s). THREE ZALLERIAN UPGRADES the diagram simplifies (deliver the first two, keep the third for questions): (1) Gate 2 isn’t belief-EVALUATION — it’s predisposition-matching via cues, and it needs contextual information to run: people resist uncongenial messages “only to the extent that they possess the contextual information necessary to perceive a relationship between the message and their predispositions” (p. 44). So the gate FAILS OPEN for low-context receivers — the inattentive “‘mistakenly’ accept” what they encounter. Gate 2 checks tribe, not truth — which is exactly why the premise-checking toolkit from the previous session matters. (2) The SAMPLE step (now on the diagram): what passes the gates joins a pool of “considerations,” and the expressed opinion is a sample of whatever is on top at answer time — in Zaller’s model conventional “attitudes” don’t exist (p. 118). Party trick, the vacillating teacher (pp. 53–54): the SAME survey respondent, emphatically pro-government-services in June, emphatically anti- four weeks later; the survey staff assumed scrambled case IDs — the IDs were correct. (3) The gates have different widths: reception rises and acceptance falls with awareness, so the MODERATELY aware are the soft target at normal message intensity (the inverted-U, pp. 123–27) — but under saturation-level flows the LEAST aware move most. More attention is not more independence. ARISTOTLE’S FOUR GAMES (Sophistical Refutations), one breath: teaching, inquiry, examination, CONTEST — and in contentious argument the goal is not establishing anything but producing the APPEARANCE of victory. Purpose is a separate axis from truth-orientation (Frankfurt, later today): a speaker can use only true facts and still be playing contest, not inquiry. The same sentence behaves differently when the goal changes from finding out to winning — ask “what game are we playing?” before scoring an exchange.

When Opinions Change

The same three filters decide it — an opinion moves only when a challenge gets through:

	Opinion stays when…	Opinion can move when…
Attention	the challenge never reaches you	you meet the other side’s actual case
Impression	the messenger is distrusted	a source you already trust says the unexpected
Fit	your group’s position is at stake	identity stakes are low — or the case is framed in your values

Change is usually gradual: new stored ideas shift the mix your answers are drawn from (Zaller 1992).

Change earned by argument lasts; change won by shortcuts fades (Petty & Cacioppo 1986).

NARRATION: One more slide before the appeals, answering the question everyone is asking: when do people actually change their minds? Run the filters in reverse. Row one, attention: most challenges to an opinion are simply never encountered — people mostly read and hear their own side, so the first condition for change is contact with the other side’s actual case, not its caricature. Row two, impression: the same correction lands differently by messenger — it works best from a source the listener already trusts, and best of all when that source says something against its usual side (Berinsky 2017: “surprising validators”). Row three, fit: when a position has become part of group membership, “must I believe it?” sets the bar impossibly high; when identity is not at stake — most topics, most of the time — people update readily. Framing a case in the audience’s own values (moral reframing: Feinberg & Willer 2015) lowers the fit barrier without deception. THE TWO FRAGMENTS: (1) Zaller again — change is rarely a switch flipping; it is new considerations entering the pool and shifting the sampled mix. That is why real change looks like drift, and why the vacillating teacher from the previous slide’s note is not a paradox. (2) The ELM durability postulate (Petty & Cacioppo 1986): change from actually processing arguments persists, resists counter-persuasion, and predicts behavior; change from shortcuts (confidence, likability, repetition) decays and needs topping up — which is why advertising never stops. EVIDENCE GUARDRAILS, have ready if asked: the “backfire effect” (corrections strengthening the false belief) mostly FAILS to replicate — corrections usually move people modestly in the right direction (Wood & Porter 2019); do not teach “facts make it worse.” Average persuasive effects of campaign contact in general elections are near ZERO (Kalla & Broockman 2018) — locked identities, closed filters — yet ten-minute perspective-taking conversations produced durable attitude change on transgender rights (Broockman & Kalla 2016): non-judgmental exchange, personal narrative, low identity threat. Persuasion is rare where identity is engaged and possible where it is not — that asymmetry IS the slide. BRIDGE, one sentence: the appeals, next, are the levers that work these filters from the speaker’s side.

The Three Appeals

Ethos · Pathos · Logos

Rhetoric — “the faculty of observing, in any given case, the available means of persuasion” (Rhetoric I.2, 1355b).

In plain terms: whatever the topic, notice everything that could move this audience.

Aristotle’s Rhetoric — the first map of persuasion:

Ethos — trust in the speaker: credibility, authority.
Pathos — the feelings of the audience.
Logos — the argument: premises and links.

Persuasion always mixes all three:

the fallacy families are these appeals abused

Raphael, The School of Athens (detail, c. 1510) — Plato in gray, Aristotle in color. Public domain.

THE IMAGE IS THE ARGUMENT, say it in one breath: Plato (gray, pointing up to ideal forms) wrote philosophy’s great attack on rhetoric — the Gorgias calls it mere flattery; his student (in color, palm pressed toward the world, the Ethics under his arm) answered with the first honest MAP of persuasion instead of a ban on it. That is this session’s stance too: map it, don’t just denounce it. The ancient map, and why it matters for persuasion: every message rides some blend of speaker-trust, feeling, and argument — so every filter-attack from the Process section has a classical name. Ethos abused = misused sources; pathos abused = emotional appeals; logos abused = faulty reasoning and distorted claims. The next three slides give each appeal its real conceptual job — this slide is only the map. LINEAGE, verbally (the second portrait): Aristotle also wrote the first fallacy catalogue (the Sophistical Refutations — equivocation is already on his list). Schopenhauer (c. 1831) wrote the cynic’s manual — 38 stratagems for winning “whether one is in the right or the wrong — per fas et nefas” (Saunders’ 1896 translation, p. 4; “by fair means or foul” is the customary gloss). The last trick, XXXVIII: “become personal, insulting, rude” (p. 45) — the fallacies read backwards as a user’s manual. And in 1937 the Institute for Propaganda Analysis taught American schoolchildren seven devices — classroom inoculation a quarter-century before McGuire named the theory; their seven map straight onto our families (name-calling, testimonial, bandwagon, glittering generalities = platitudes, card stacking = the one-sided premise menu). THE CAPTION, unpacked (SEP on Rhetoric I.1–2): Aristotle DEFENDS rhetoric — it is dialectic’s counterpart, and even a speaker with exact knowledge cannot TEACH a public audience, so what is just and true needs rhetorical help or it loses to what merely sounds good. And his defense of studying persuasion-for-both-sides is the course’s inoculation logic, 2,300 years early: command of both sides is not a license to argue both — it is what “helps to detect what goes wrong in the opponent’s arguments.”

Ethos

The question it answers: “why should I listen to you?”

Borrowed trust: the white coat, the title.
Performed trust: sounding confident and composed.
Relational trust: “I’m one of you.”
Counter-ethos: make the other side untrustworthy.

Trust opens every filter — it decides whose evidence even gets heard.

Ethos with its real job: not “the speaker has credentials” but the CONSTRUCTION of a speaker the audience will listen to — knowledgeable, morally trustworthy, similar, committed to their interests, composed. ARISTOTLE’S OWN RECIPE (Rhetoric II.1, via SEP): display practical intelligence, good character, and good will toward the audience — and the trust must arise FROM THE SPEECH ITSELF, not from prior reputation; he adds, coolly, that the speaker need not actually BE virtuous, only appear so in the speaking. In other words: “performed ethos” is not a modern cynical add-on — it is in the original manual, flagged by its author. The four forms, one example each: BORROWED — the Lab Coat from the source family (the coat is rented, the survey real, the question wrong); PERFORMED — fluency and composure read as competence (hold this one: it is the key to the Reagan retrace later); RELATIONAL — “one of us” (the identity half of bandwagon); COUNTER-ETHOS — the ad hominem re-described from the sender’s side: kill the messenger’s credibility and the message dies unheard, which in Zaller’s terms is manufacturing a cueing message. HONEST vs. COUNTERFEIT: earned, in-field expertise with methods shown vs. rented coats and performed confidence — the earlier expert-testimony criteria are the test. ETHOS AT ZERO CONTENT (Todorov, in Kahneman 2011, ch. 8): competence judgments from one-tenth-of-a-second exposures to candidates’ faces predicted about 70% of US Senate, House, and governor races — replicated in several countries, including Mexico — and the effect was about three times larger among low-information, TV-heavy voters. The instrument works before a single claim is made — and, as with Zaller’s gates, it works hardest on whoever checks least. And recall the McGuire result from the source-family session (via Zaller 1992, p. 45): a high-credibility source makes the same message look “even more grammatical” — source cues corrupt the logical judgments themselves.

Pathos

The question it answers: “what should matter to me right now?”

Each emotion is a hidden claim:

Fear: “the danger is near.”
Anger: “someone is to blame.”
Pity: “someone needs help.”
Pride: “this is what we do.”
Humor: “relax — we’re friends here.”

Pathos upgraded from “emotion contaminates reason” to its real job — and this IS Aristotle’s own definition (Rhetoric II.1, via SEP — Rapp 2022): the emotions are “those things due to which people, by undergoing a change, DIFFER IN THEIR JUDGEMENTS.” Not noise around judgment — a change in the judge. He then analyzes them pair by pair (anger, fear, pity, shame, envy…) precisely as levers a speaker can work. Emotion is part of JUDGMENT — it decides what receives attention, which facts appear important, how risks are weighed, what is remembered, whether action feels urgent, and who counts as “us.” Each emotion carries a MESSAGE — read the five aloud slowly; each one is a claim the audience never gets to examine as a claim. This retroactively deepens the feelings family from the fallacies sessions: fear and pity were never irrational noise — they were unexamined CLAIMS about immediacy and desert, offered as the whole case. The honest form has evidence behind the emotion: fear appeals persuade legitimately — and most effectively — when the stakes are real AND the message includes efficacy, something the audience can do (Tannenbaum et al. 2015, meta-analysis). The test is never “was there emotion?” — it is “was the feeling offered as the whole case?” (the fallacies session’s criterion, unchanged). HUMOR is the entry the catalogue never had: it performs ethos (composure) while delivering pathos (a shared social moment, defenses down) — hold it for the retrace.

Logos

The question it answers: “why does this seem to follow?”

The honest version: valid form, true premises — the ladder.
The persuasive shortcuts: stories · examples · numbers · framing.
The oldest trick: leave a gap — the listener fills it (enthymeme).
What you conclude yourself, you rarely doubt.

The counterfeit: sounds like reasoning, offers no checkable claim — “studies show…”

Logos wider than formal validity: audiences are moved by causal stories, examples and analogies, statistics, framing, contrasts, narratives — each can carry a genuine argument or merely LOOK like one. THE ENTHYMEME is the slide’s center, and it is the course’s oldest skill viewed from the sender’s side: the hidden-premise hunt (from the argument-reconstruction sessions) was practice at EXPOSING what persuaders leave for the audience to supply. When the listener completes the argument, the conclusion arrives with the feeling of self-discovery — people rarely argue against their own inference. Paltering (the intent session’s weaponized enthymeme) is this engine used deliberately: assert only true premises, let the listener build the falsehood. ARISTOTLE ON THE ENTHYMEME (via SEP — Rapp 2022): the enthymeme is “the body of persuasion” — people are most easily persuaded “when they suppose something to have been proven” — and it is built from ENDOXA, opinions the audience already accepts; that is exactly why supplying the missing premise feels effortless: it was already yours. His practical rule: keep enthymemes SHORT — “the members of a jury or assembly are not accustomed to following a longer chain of inferences” — which is the previous session’s cost-of-checking economics, stated in the fourth century BC. COUNTERFEIT logos = Vaughn’s proof surrogates (“studies show,” “experts agree,” “the data are clear” — which studies? which data?): the grammar of evidence with no evidence attached. The defense is unchanged since week two: write the argument out, premises and conclusion, and see what is actually there.

The Instruments, Used and Abused

Every appeal has an honest use — the trouble is persuasion “through rhetoric alone” (Vaughn 2019):

The appeal	Used honestly	The counterfeit
Ethos — “trust me”	a cardiologist shows how she knows	the rented white coat in an ad
Pathos — “feel this”	a real danger — plus what to do	fear or pity doing all the work
Logos — “it follows”	reasons you can check, step by step	“studies show…” — which studies?

Try it (2 min, in pairs) — one sentence from an invented ad:

“As a doctor and a mother, I say: act now, before our children pay — the data are clear.”

1. Find “trust me,” “feel this,” and “it follows” in the sentence.

2. What does it want you to conclude? What did it actually prove?

The reveal: all three instruments — and nothing you could check.

The consolidation slide: the whole fallacies catalogue, re-sorted under the three appeals — and the honest column is the point, delivered with the textbook behind it. VAUGHN’S CRITERION (Ch. 5, the assigned pages): rhetoric is not inherently wrong — the worry is persuasion “through rhetoric alone” (p. 152); his spectrum runs honest appeal → fallacy (“the semblance of arguments”) → pure persuader (“nonargument persuaders,” p. 152); quotable, p. 174: “Rhetoric presents us with a large repertoire of techniques for influencing hearts and minds through emotion — while proving nothing through reason.” ARISTOTLE’S SIDE (Rhetoric I.2, the recommended reading): the three appeals are the pisteis — legitimate modes of proof; rhetoric is dialectic’s counterpart, not its corruption. THE CELLS, unpacked in one breath each: the cardiologist is in-field expertise with METHODS shown (the expert-testimony criteria); “plus what to do” is the efficacy condition — fear appeals persuade legitimately when stakes are real AND the message says what to do about them (Tannenbaum et al. 2015); “check, step by step” is the ladder — valid form, true premises; and the white coat, the whole-case feeling, and the sourceless “studies show” are the Lab Coat, the feelings family, and the proof surrogate, all met before. The quoted tags (trust me / feel this / it follows) are the plain handles — use them for the rest of the session. SOLUTION, delivered after the pairs report. TASK 1 — where each handle lives (cold-call one phrase each): “as a doctor” → TRUST ME (borrowed ethos — and no field is ever named); “and a mother” → TRUST ME + FEEL THIS (identity and sympathy in one word); “act now, before our children pay” → FEEL THIS (fear, with a clock on it); “the data are clear” → IT FOLLOWS (a logos costume — WHICH data?). TASK 2 — the two answers, side by side. What it wants you to conclude: “our children are in real danger, and we must act immediately.” What it actually proved: nothing — no danger named, no evidence shown, no action specified. The gap between those two answers IS the enthymeme: the audience builds the argument themselves (doctors understand danger; mothers protect children; therefore act) from premises the ad never stated — which is why the conclusion feels self-evident. THE REVEAL, say it slowly over the last fragment: one sentence, all three instruments, and not one checkable claim. That is Question 2 of the defense (coming): what claim is actually on offer? The dissection exercise later runs this same skill on three real specimens.

Working Together

Interlocking Appeals

The appeals aren’t three separate scores — each changes the others:

Trust someone → you believe their numbers. (ethos → logos)
Fear something → its facts feel bigger. (pathos → logos)
A good argument → “she’s smart.” (logos → ethos)
A shared laugh → “he’s one of us.” (pathos → ethos)
Hear it often → it feels familiar — and familiar feels true.

So don’t judge a message one appeal at a time — trace the sequence.

The real story of persuasion: people do not process three independent appeals and add the scores — each appeal changes how the others are interpreted. Walk the loop: ethos gates the evidence (an untrusted source’s data never get weighed); pathos weights the facts (the frightening one counts double); logos gives the trust and the feeling somewhere to arrive — and then it runs backwards: “she made a strong argument, so she must be competent” (logos manufacturing ethos), humor performing composure (pathos manufacturing ethos), repetition breeding familiarity that is read as BOTH credibility and plausibility (the illusory-truth literature from the previous session: Hasher 1977; Fazio et al. 2015). This is why one-instrument defenses fail and why the defense section asks QUESTIONS instead of banning appeals. Bridge: now retrace the opening joke with all of this on the table.

The Joke, Retraced

The opening puzzle — retraced with the tools you now have:

1. Attention	The oldest president is asked, point-blank: can you function?
2. “Trust me” (ethos)	The joke shows composure — the very quality the question doubted.
3. “Feel this” (pathos)	Humor: “relax, we’re friends” — the hall laughs, even Mondale.
4. “It follows” (logos) — the premise you supply	“A man this calm under fire has stamina to spare.”
5. What would decide it	Actual evidence of stamina — which the joke never gave.

The puzzle, solved: the reply didn’t argue — it made you build the argument.

The habit to keep: after the laugh, ask — was the question answered?

The payoff slide — the opening puzzle, solved with the tools the room now owns, and the WORKED EXAMPLE for the exercise coming: its five rows are exactly the exercise’s five steps, and rows 2–4 are the drill’s three handles run on a real specimen. Row 2 is the deepest point — say it slowly: the joke does not just charm, it SHOWS composure under pressure — the reply’s form is itself the evidence for the quality Trewhitt doubted. That is the Ethos slide’s “performed trust” at the highest stakes. Row 3 is the Pathos slide’s humor line (“relax — we’re friends here”) happening on tape: defenses down, a shared moment — and Mondale laughing shows even the victim inside it. Row 4 is the Logos slide’s gap: no stamina claim is ever asserted; the audience supplies “he can clearly handle pressure” as their own discovery — which is why it stuck (the drill’s task 2, again: conclude much, prove nothing). Row 5 is the course’s oldest question — what would decide it — and the answer the joke never gave. Guard the criterion: humor in a debate is NOT a fallacy (nothing is argued badly); the move is a joke OFFERED AS the answer — and the audience, not the speaker, closes the question. GIVE THE ROOM THE CLASSICAL NAME, verbally: a charming subject-change is the RED HERRING (ignoratio elenchi’s popular form) — and the joke is its most elegant version: after the laugh, the subject has moved. Not a catalogue entry — a conduct trick, like the gallop. And the same-claim follow-up, from the fallacies session’s test: did the reply address the SAME question Trewhitt asked? No — it changed the task entirely. Keep the claim modest: what the clip documents is that night’s exchange; the election had many causes.

Whose Hands?

Ignorance vs. Intent

The claim returns — “the cafeteria scored 42” — from two very different senders:

	shared believing it	spread knowing it’s false
The name	misinformation	disinformation
The failure	an honest bad premise	a deliberate lie

Working definitions — fields draw these lines differently.

Fallacies and floods, used deliberately, are tools to gain power.

The same catalogue shows exactly how these tactics work — learning it is the defense.

Open with the section’s point: the SAME persuasive mechanisms serve an honest teacher, a mistaken citizen, a liar, and a propagandist — the appeals never tell you which sender you face; that must be asked separately. Then the canonical distinction, finally named: MISinformation = false content shared in good faith (the sender is a victim too); DISinformation = false content spread with intent (the sender is the attacker). Same claim, same damage — different moral ledger, and different fix (correct the ignorant; expose the cunning). Walk it on the specimen: sender A repeats “42” from a friend and believes it — misinformation; sender B has read the report showing 92 and posts “42” anyway — disinformation. Say the working-definitions line aloud: fields draw these boundaries differently; these are the course’s working definitions. A THIRD orientation to truth — neither honest error nor lying — gets its own slide next; park it if a student jumps ahead. PALTERING (Rogers et al. 2017) is the weaponized ENTHYMEME: the palterer asserts only true premises and lets the LISTENER add the hidden premise and draw the false conclusion — you commit the fallacy for them (Clinton’s “there IS not a sexual relationship — that is accurate”). The antidote is the earlier hidden-premise skill: write out the implied premise and check it; the deception lives exactly where the enthymeme lives. MERCIER & SPERBER explain the formal session’s core phenomenon — the conclusion-first feeling: reasoning evolved to produce justifications and persuade, so myside bias generates cherry-picked premise sets that feel complete. Logical moral: an argument can be perfectly VALID and still rigged by premise selection — validity checks the link, never the menu. THE OTHER HALF, so the citation stays honest: that is the PRODUCTION face of reasoning — biased and lazy; the EVALUATION of other people’s arguments is demanding and fairly objective, and Mercier & Sperber explicitly reject the cynical-only reading of their book. (The Trouche receipt — people rejecting their OWN argument when disguised as a stranger’s — lives on the Who Checks You slide near the close.) VAUGHN’s pairing (Ch. 5, “Fallacies and Persuaders”) completes it: fallacies persuade because they mimic the FORM of argument; pure persuaders (innuendo, ridicule, euphemism) don’t even bother. Tone guard still applies: much fallacious reasoning is sincere — this slide is about the subset that isn’t. VAUGHN’S PERSUADERS, itemized if asked — the argument-FREE toolkit, moves too small for the catalogue because no argument is ever offered: euphemism/dysphemism (“enhanced interrogation” vs. “torture”), innuendo (“I’m not saying he’s corrupt, but…”), weasel words (“may,” “up to,” “some say”), proof surrogates (“studies show” — which studies? the Lab Coat’s cousin). No inference exists to check, so the only defense is NAMING the device. And the “user’s manual” line is literal: Schopenhauer wrote it — the 38 stratagems named on the Ethos map earlier.

Indifference to Truth

The third sender never opened the report — said 42 because it worked (Frankfurt 2005):

	relationship to truth
The honest speaker	tries to describe reality accurately.
The liar	knows the truth matters — and works to conceal it.
The bullshitter	says whatever works; truth is incidental.

The diagnostic: is the speaker trying to establish the truth, conceal it — or just produce an effect?

The guard: a false claim doesn’t prove lying — and a true one doesn’t prove care.

Frankfurt’s contribution, and why it earns a slide: the mis/dis ladder sorts speakers by whether they know the claim is false — but the bullshitter isn’t ORIENTED to truth at all. The liar still plays the truth game (tracks truth-values in order to negate them); the bullshitter exits the game — so the output can be false, exaggerated, misleading, or accidentally TRUE. Truth doesn’t make the process responsible: a claim produced by guessing or image-management is bullshit even when it happens to land on a fact. Tautologies and platitudes are bullshit’s small change — “true” because they say nothing. KEEP IT SEPARATE FROM THE GALLOP, say this explicitly: the gallop is a diagnosis of PACE and structure; bullshit is a diagnosis of ORIENTATION. A galloper can sincerely believe every claim they fire; a bullshitter needs only one. A gallop turns Frankfurtian when the claims are selected for rhetorical usefulness rather than support. This is the fallacy-fallacy guard again: a fast, weak argument does not prove the speaker is lying. THE SECOND FRAGMENT is the two-axis lesson: truth-value (true / false / uncertain) and orientation (investigating / deceiving / indifferent) are SEPARATE dimensions — you cannot read intent off truth-value in either direction. Quick drill if time (2 min): give four statements, classify each on both axes. WHY BULLSHIT PROLIFERATES (Frankfurt): people feel obliged to have opinions on everything public — panels, feeds, seminar bluffing. Our extension (not Frankfurt’s): generative AI cuts the cost of polished assertion, not the cost of verification. DISCUSSION CLOSER if time: Frankfurt argues bullshit is the GREATER enemy of truth than lying — the liar still acknowledges truth’s authority; the bullshitter ignores it. Present it as his philosophical argument, not an empirical finding. CLOSE THE LOOP with the opening specimen: Frankfurt’s own example is the political orator whose descriptions are instruments of SELF-PRESENTATION — projecting patriotism, seriousness, moral elevation. The retrace already named the Reagan reply’s orientation: it neither lies nor informs — no checkable claim, an image produced. The third orientation, at its most charming.

The Defense

Three Questions

Everything today distills into three habits:

Was the question answered?
What did it actually prove — is there a checkable claim?
Which premise did you supply?

Ask them after the charm has worked — that is the discipline.

The defense, distilled from the session’s own moves — walk the sources aloud. Question 1 is the puzzle’s habit: after the laugh, was Trewhitt’s question answered? Question 2 is the drill’s task 2 and the counterfeit-logos test: conclude much, prove nothing — find the checkable claim, or name its absence. Question 3 is the enthymeme turned into a mirror: the most persuasive premise in any message is the one YOU supplied — say it out loud and check it; this is the hidden-premise skill from the argument sessions, pointed at yourself. The fragment carries the discipline: these questions cost nothing to ask BEFORE a message charms you — the skill is asking them AFTER, when the laugh has landed and the conclusion already feels like yours. A FOURTH question for when the sender matters (from Whose Hands): would this sender care if it were false — and whose interest does the selection serve? THE STING in Question 3 sets up the next slide: if the premise was yours, who checks the argument you built? Not you — next.

Who Checks You?

The joke worked because you built the argument. So — who checks it?

You? No — everyone finds their own reasoning convincing.
A friend who agrees? No — they’ll just nod along.
Someone who disagrees? Yes — they’ll find the weak spot in seconds.

The evidence: alone, about 1 in 10 people solve a logic puzzle — arguing it out in a group, 8 in 10 (Mercier & Sperber 2017).

The last defense: say your reasoning out loud to someone who disagrees.

The session’s final turn of the screw, and the chain to say aloud: the enthymeme plants conclusions INSIDE you (Logos: what you conclude yourself, you rarely doubt; the Retrace: the reply made YOU build the argument) — so your own conclusions are exactly where your checking is weakest. Walk the three candidate checkers: YOU fail because self-checking is exactly what the enthymeme defeats; the AGREEING friend fails because agreement supplies no resistance — groups of the like-minded don’t converge on truth, they drift further in the direction they already lean; the DISAGREEING checker works because evaluating someone else’s argument is the one reasoning skill humans are demonstrably good at. The lab receipt (Trouche, in Mercier & Sperber 2017, pp. 231–233): people tricked into judging their OWN argument as a stranger’s rejected it more than half the time — we grade our own reasons on a curve, and other people’s honestly. That asymmetry is the book’s thesis: reason misfires in its abnormal habitat (solo, or an echo chamber) and performs remarkably in its evolved one (genuine back-and-forth). The numbers: the Wason selection task runs ~15–25% correct for individuals; discussion groups reach 80% (Moshman & Geil, reported in M&S, pp. 264–265) — their line: like getting sprinters to run 100 meters in five seconds by making them run together. Mock juries beat individual jurors the same way (p. 272). THE CONDITION carries the slide, don’t soften it: the gain needs DIFFERENT views and a COMMON goal — groups of the like-minded don’t converge on truth, they polarize (pp. 247–250). That is the difference between a seminar and an echo chamber, and it is why every exercise in this unit ran in pairs. Keynes for the solo half, quotable: “It is astonishing what foolish things one can temporarily believe if one thinks too long alone” (p. 250); Bertillon (the self-sealing Dreyfus expert, from the circular-reasoning session) is its catastrophic case. Hand-off: the dissection exercise next, then the conclusion; WHY solo checking is so weak is a question for the cognitive-biases reading (Kahneman, ch. 1). SEMINAR BEAT, the Zaller tension, if the room is strong: Zaller’s receivers never argue back — his model “makes no allowance for citizens to think, reason, or deliberate” (p. 45), and his data say resistance depends on externally supplied counter-messages, not internal resources (p. 267). Mercier & Sperber’s reply is this slide: reasoning was never designed for the lone receiver — it shines in exactly the exchange Zaller’s surveys never observe. Both agree the lone receiver is weak; keep the counter-flow alive.

Exercise — The Dissection

In pairs (6 min) — pick one specimen:

1. “You’re not you when you’re hungry.” (a chocolate ad)

2. “Trust me — I’ve used it for years, and it changed my life.” (an influencer)

3. “Everyone is switching — don’t be the last.” (a phone ad)

Your task: fill the joke’s five rows for your specimen — attention · trust me · feel this · the premise you supply · what would decide it.

Debrief: which specimen carried the least argument — and persuaded the most?

The Retrace, run by the students on fresh specimens — say it plainly: “do to your specimen exactly what we did to the joke; same five rows.” ANSWER KEY, five rows each. (1) THE CHOCOLATE AD: attention — the absurd transformation (a friend becomes a diva when hungry); trust me — only brand familiarity, no one vouches for anything; feel this — humor: “relax, we’re friends”; the premise you supply — “hunger makes me not myself → this bar fixes hunger → buy it” (the middle premise does all the work); decide it — would any snack do the same? (2) THE INFLUENCER: attention — direct address, a face you follow; trust me — the strongest row: relational (“I’m one of you”) plus performed sincerity, yet zero credentials and, usually, an undisclosed payment; feel this — aspiration: “changed my life” sells a future self; the premise you supply — “it worked for them, so it will work for me”; decide it — what exactly changed, measured how — and were they paid to say it? (3) THE PHONE AD: attention — urgency (“don’t be the last”); trust me — the crowd as the authority, bandwagon = relational ethos at scale (the popularity family from the fallacies sessions); feel this — pride and fear of being left out in one clause; the premise you supply — “if everyone is switching, it must be better”; decide it — is it actually better? A count of buyers is not a measure of quality. DEBRIEF TARGET: the specimen with the LEAST argument often persuades MOST — the session’s thesis, demonstrated by the students themselves. All three offer NO checkable claim — Question 2 of the defense returns empty on every one; make the room notice. TIME SAVER: assign one specimen per third of the room instead of free choice; debrief one row per specimen.

Conclusion

How does a message move a mind?

Persuasion begins by winning attention.
Ethos decides whom we trust; pathos, what feels urgent.
Logos supplies the path — sometimes genuine, sometimes counterfeit.
The same techniques serve the honest, the mistaken, the liar, the indifferent.
The defense: three questions — answered? · proved? · which premise was yours?

Homework: find one ad or speech — trace the sequence, then run the three questions.

The one-slide recap, read as the session’s story, close to verbatim: persuasion begins by winning attention; ethos determines whom we trust; pathos determines what feels important; logos supplies a path — sometimes genuine, sometimes counterfeit — from those judgments to a conclusion; and critical thinking interrupts it by reconstructing the claim, the missing premises, and the speaker’s purpose. Line to close on, verbatim: “The gates are easy to grease from the outside — WHY they are just as easy to grease from inside your own head is a question for the cognitive-biases reading — Kahneman, chapter one.” (Point at the reading, do not resolve it. The next taught session is correlation and causation.) Homework: one ad or speech — trace the sequence in writing (attention, trust, feeling, the supplied premise, what would decide it), then the three questions. Alternative: the Bad News game (getbadnews.com, ~15 minutes) if they haven’t played it yet.

References

Aristotle. Rhetoric. (any scholarly edition)
Frankfurt, H. G. (2005). On bullshit. Princeton University Press.
Gilovich, T. (1991). How we know what isn’t so: The fallibility of human reason in everyday life. Free Press.
Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.
Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480–498.
Mercier, H., & Sperber, D. (2017). The enigma of reason. Harvard University Press.
Petty, R. E., & Cacioppo, J. T. (1984). The effects of involvement on responses to argument quantity and quality: Central and peripheral routes to persuasion. Journal of Personality and Social Psychology, 46(1), 69–81.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 19, pp. 123–205). Academic Press.
Vaughn, L. (2019). The power of critical thinking: Effective reasoning about ordinary and extraordinary claims (6th ed.). Oxford University Press.
Zaller, J. R. (1992). The nature and origins of mass opinion. Cambridge University Press.