PHIL 252 — Unit 10: Putting Critical Thinking into Practice

The final unit integrates every tool covered in the course. It answers three questions: (1) How do we reason well in the real world? (2) What does a complete argument evaluation look like in practice? (3) How do we protect ourselves from bullshit in a data-driven world?

Learning outcomes:

  • Apply course concepts to everyday critical thinking contexts
  • Explain and reflect on approaches and attitudes about critical thinking
  • Describe and evaluate arguments from real media

Part A — Chapter 20: Putting Critical Thinking into Practice (Dayton & Rodier)

20.1 Returning to Inductive Strength

Most real-world knowledge is merely probable, not certain. Inductive reasoning is limited because humans are not omniscient and cannot know the future. Our beliefs form a network of support relations — not isolated claims. The fact that beliefs are “merely probable” doesn’t mean you can believe whatever you like: beliefs are constrained by the entire network of things you know.

We organize beliefs into schemas and scripts (stories, goal-oriented plans, stereotypical situations). These schemas ground our inferences.

20.2 Material Inferences

Material inferences are defeasible inferences that depend on informal patterns — typical features, probable signs, reasonable assumptions. They are:

  • Neither universal nor necessary
  • Defeated by additional information
  • Ineliminable from human reasoning

Four types: motivational (infer motive) · feature (infer from typical property) · resultative (infer consequence) · functional (infer purpose)

See MaterialInferences for the full breakdown.

20.3 The Three-Step Argument Analysis Procedure

StepWhat You DoWhat You’re Testing
1. Identify / Clarify / DistinguishPut argument in standard form; make implicit premises explicitAre the claims what they appear to be?
2. Dialectical AcceptabilityAre premises true, likely, and relevant?PREMISES
3. Logical ConnectionDo premises give sufficient rational grounds?CONNECTION between P and C

See ArgumentAnalysisProcedure for full detail, cogency conditions, and the St. Albert Gazette worked example.

Cogency conditions (review):

  1. Premises rationally acceptable to a reasonable audience
  2. Premises make a rationally grounded connection to the conclusion
  3. Premises provide sufficient rational grounds

Walton’s 5 conditions for fallacies: argument · falls short of correctness · dialogue context · semblance of correctness · poses problem to the dialogue goal

20.4 Attitudes of Critical Thinking

  • Fallacies are symptoms of bad reasoning — use them as diagnostic categories, not gotcha labels
  • Be charitable: take even irritating arguments seriously as claims for reasonableness
  • Taking cognitive ownership: evaluating an argument means taking it on as your own and asking whether you would endorse its structure
  • Ask: is anything good in this argument? Can any parts survive if the fallacies are isolated?

Part B — Chapter 21: Fallacy Round-Up (Dayton & Rodier)

A complete reference table of all fallacy categories covered in the course.

21.1 Fallacies of Ambiguity (Ch. 14)

FallacyCore Idea
EquivocationKeyword used in two senses; argument depends on the shift
AmphibolyStructural ambiguity in a sentence’s grammar
AccentMeaning shifts depending on which word is stressed
CompositionFrom parts → whole (invalid); from member → class (invalid)
DivisionFrom whole → parts (invalid); from class → member (invalid)
HypostatizationAbstract word or metaphor treated as a concrete thing

See FallaciesOfAmbiguity

21.2 Fallacies of Emotional Bias (Ch. 15)

FallacyCore Idea
Ad hominem (personal attack)Reject claim by attacking the person
AbuseName-calling to redirect attention from the issue
Poisoning the wellAttack person’s motivation rather than their argument
Tu quoque”Look who’s talking” — person’s behaviour undermines their argument
Mob appealSway belief via theatrical language or group-based interests
Appeal to pityEvoke pity to cause assent
Appeal to force/fearThreats of force to cause acceptance
Two wrongs make a rightJustify behaviour by asserting the other party would do the same

See FallaciesOfEmotionalBias

21.3 Fallacies of Expertise (Ch. 16)

FallacyCore Idea
Appeal to authorityAccept claim just because an expert says so (without evaluating the expertise)
Snob appealBelieve this and you’ll join an exclusive superior group
Appeal to traditionPast practice is reason enough to continue it
Appeal to natureNatural = good; unnatural = bad
Appeal to anonymous authorityClaim supported by unspecified “experts”
Appeal to ignoranceFailure to disprove a claim = reason to accept it

See FallaciesOfExpertise · AppealToAuthority

21.4 Fallacies of Distorting the Facts (Ch. 17)

FallacyCore Idea
False analogyTwo things only superficially similar, or not similar in the relevant respect
False cause (family)Insufficient evidence for a causal claim
Post hoc ergo propter hocB happened after A → A caused B
Mere correlationB correlates with A → A caused B
Reversing cause and effectA causes B when actually B causes A
Spurious correlationA causes C when both are effects of B
Slippery slopeEvent must inevitably lead to consequence — no argument for inevitability
Irrelevant thesisSidestep the real issue; claim it has been settled by the diversion

See FalseCause

21.5 Fallacies of Presumption (Ch. 18)

FallacyCore Idea
Sweeping generalization (accident)Apply a general rule to a special case where it doesn’t apply
Hasty generalization (converse accident)Build a general rule from a special/unrepresentative case
BifurcationPresent a false choice; treat contraries as contradictories

See FallaciesOfPresumption

21.6 Fallacies of Evading the Facts (Ch. 19)

FallacyCore Idea
Straw personDistort opponent’s view to make it easy to knock down
Begging the questionAssume what you’re trying to prove
Question-begging epithetsSlanted language that implies what hasn’t been proved
Complex questionQuestion that presupposes the truth of what’s at issue
Special pleadingApply slanted language to others; use neutral language for yourself

See FallaciesOfEvadingTheFacts


Part C — Chapter 10: Spotting Bullshit (Bergstrom & West)

Six practical tools for identifying misleading claims, especially in media and social media contexts. See Bullshit for the expanded page.

ToolThe RuleKey Example
1. Question the sourceWho is telling me this? How do they know? What are they selling?Goop healing crystals — questions reveal lack of real expertise
2. Beware unfair comparisonsEntities must be directly comparable for the comparison to be valid”Most dangerous cities” lists — city boundary definitions skew rankings
3. If too good/bad to be true → dig to the sourceSocial media amplifies extreme claims; the most viral posts are often those that shockNBC tweet: “applications down 40%” vs actual report (applications down at 40% of schools)
4. Think in orders of magnitude (Fermi estimation)Make back-of-envelope approximations to check plausibility; being off by 50% still keeps you within tenfoldFood stamp fraud: 0.2% of budget lost, not a scandal; National Geographic 9 billion vs 9 million
5. Avoid confirmation biasTendency to notice, believe, and share information consistent with preexisting beliefsLetter-of-recommendation gender bias study: tweet showed the hypothesis, not the actual (mixed) results
6. Consider multiple hypothesesJust because someone has an explanation doesn’t mean it’s the explanationDisney stock down 2.5% attributed to Roseanne cancellation — but the drop occurred before the announcement

Online bullshit tactics:

  • Corroborate and triangulate (check multiple sources)
  • Pay attention to where information comes from
  • Dig back to the origin; read the full story, not just the headline
  • Use reverse image lookup (TinEye, Google Images)
  • Be aware of deepfakes and synthetic media
  • Use fact-checking sites: Snopes.com, PolitiFact.com, FactCheck.org
  • Illusory truth effect: the more often you see something, the more likely you are to believe it → reduce information intake; “think more, share less”

Part D — Chapter 11: Refuting Bullshit (Bergstrom & West)

Three tools for calling out bullshit once you’ve spotted it.

ConceptDefinitionExample
Calling bullshitA performative utterance in which one repudiates something objectionable; scope broader than bullshit alone (can call bullshit on lies, treachery, injustice)Finding a reputable source that refutes a rumour; citing Snopes
Reductio ad absurdumRefuting a claim by deriving an absurdity from its logical extensionWomen’s running times: if the trend continued, women would eventually run 100m in 0 seconds
Null modelA model that shows what we would observe in a very simple system where not much is going on; test whether alternate explanations fit the same dataThe fastest runner isn’t necessarily older — it’s the one in the largest sample; age effect disappears when sample size is controlled

Part E — Chapter 11: Categorical Equivalence Review (Dayton & Rodier)

A review of immediate inference operations and their equivalences, with the concept of distribution made explicit.

Distribution

A term is distributed if the statement makes a claim about every member of the class referred to by that term.

StatementSubject distributedPredicate distributed
A (All S are P)YesNo
E (No S are P)YesYes
I (Some S are P)NoNo
O (Some S are not P)NoNo*

*Traditional logic distributes the predicate of O, but modern categorical logic takes neither term as distributed in O statements (to avoid existential import problems).

Equivalence Summary

OperationHowEquivalent for
ConversionSwitch S and PE and I only
ContrapositionSwitch S and P; negate both with “non-”A and O only
ObversionChange quality (all/no, some/some are not) + replace P with complementAll four (always valid)
RelationPairProperty
ContradictionA↔O, E↔IOpposite truth values; cannot both be true, cannot both be false
ContraryA vs. ECannot both be true; can both be false (requires non-empty classes)
SubcontraryI vs. OCannot both be false; can both be true (requires non-empty classes)
SubalternationA→I, E→OIf universal is true, particular must be true (requires non-empty classes)

Square of Opposition: Only contradiction holds in modern categorical logic. Contrary, subcontrary, and subalternation require the assumption that classes are not empty.

See ImmediateInference for the full breakdown and worked examples.


Key Distinctions for This Unit

ConceptKey Distinction
Material vs. formal inferenceMaterial = defeasible / informal; formal = necessary / valid
Confirmation bias vs. illusory truth effectBias = you seek confirming info; illusory truth = repeated exposure makes false things feel true
Calling bullshit vs. lyingLying = knows the truth and contradicts it; bullshitting = indifferent to truth
Contrary vs. contradictoryContradictory: one must be true; Contrary: both can be false
Dialectical acceptability vs. truthDA = stands up to a reasonable audience; truth = corresponds to fact

Mindmap

mindmap
  root((Unit 10<br/>Putting CT<br/>into Practice))
    Material Inferences
      Motivational
      Feature
      Resultative
      Functional
      Defeasible
    3-Step Analysis
      1 Identify & Clarify
      2 Dialectical Acceptability
      3 Logical Connection
      Cogency Conditions
      Walton 5 Conditions
    Fallacy Round-Up
      Ambiguity Ch14
      Emotional Bias Ch15
      Expertise Ch16
      Distorting Facts Ch17
      Presumption Ch18
      Evading Facts Ch19
    Spotting Bullshit
      Question the Source
      Unfair Comparisons
      Too Good to Be True
      Fermi Estimation
      Confirmation Bias
      Multiple Hypotheses
    Refuting Bullshit
      Calling Bullshit
      Reductio ad Absurdum
      Null Model
    Categorical Equivalence
      Distribution
      Conversion E and I
      Contraposition A and O
      Obversion All
      Square of Opposition

(diagram saved)