Fallacies of Presumption
Fallacies of presumption are unsound because of the unfounded or unproven assumptions embedded in them.
By smuggling in hidden assumptions, these fallacies give the impression of valid arguments while failing to actually support the conclusion. They are pervasive because human beings have non-rational motives for protecting certain beliefs — social conservatism, confirmation bias, and false consensus effects all reinforce presumptuous reasoning.
How It Appears Per Course
PHIL 252
Introduced in Ch. 18 (Unit 9). Specifically deals with how generalizations and classifications can misconstrue the relevant features of claims — either by applying a rule too broadly (sweeping), building a rule from too little evidence (hasty), or presenting a false either-or choice (bifurcation).
Background: What Is a Generalization?
A generalization is a statement made about a property of all or most members of a class.
Generalizations drive much of our reasoning (laws, rules, science). They are typically true in general but have boundary conditions — circumstances under which the rule does not apply. The fallacies of presumption arise from mishandling those boundary conditions.
The Three Fallacies
1. Sweeping Generalization (Fallacy of Accident)
Committed when an argument that depends on the application of a generalization or rule to a particular case is improper because a special circumstance (accident) makes the rule inapplicable to that particular case.
- Direction: takes a general rule → tries to apply it to a specific case → the specific case has a special feature that blocks the inference.
- The conclusion may not be false per se — the problem is that you cannot correctly draw that inference from the information you have.
- Examples:
- “Everyone has a right to advance their ideas, so judges have a right to use official positions to advance their religious views.” (Special circumstance: judges hold public office with a responsibility to remain neutral.)
- “Everyone has a right to own property, so even though Mary is a violent psychopath, we cannot take away her weapon collection.” (Special circumstance: the nature of the weapons.)
- “Cross-country skiing is healthy, so George should do more of it for his heart condition.” (Special circumstance: George has a pre-existing heart condition.)
2. Hasty Generalization (Converse Accident / Secundum Quid)
Committed when an argument develops a general rule in an improper way because it reasons from a special case (accident) to a general rule.
- Direction: takes a specific case (often exceptional or atypical) → draws a general rule from it → the specific case is an accident that doesn’t support the generalization.
- Also called over-generalization. Often arises from samples that are too small or biased.
- “Just because” test: just because there was a special case doesn’t mean it’s statistically common enough to ground a rule.
- Examples:
- Polling only coastal states (California, Florida, Maine) to conclude that “55% of Americans live near the ocean.”
- “Mary couldn’t escape her car because of her seatbelt, so seatbelts are more dangerous than going without.”
- “During the war, enemy espionage was exposed by tapping wires, so authorities should tap all suspicious persons’ phones.” (Wartime is a special circumstance; civil rights can’t be swept away in peacetime.)
3. Fallacy of Bifurcation (Either-Or / False Dichotomy)
When an arguer treats a distinction of classification as exclusive and exhaustive of the possibilities when in fact other alternatives exist. This fallacy confuses contraries with contradictories.
- Also called: false dichotomy, excluded middle, false dilemma, either-or fallacy.
- Argument form: “There are only two options, A and B. B is false. Therefore A is true.” — but other options (C, D…) may exist.
- Key distinction:
- Contradictories: one must be true, the other false. “Alive or not alive” — no middle ground.
- Contraries: cannot both be true, but can both be false. “Wednesday or Thursday” — could be Sunday.
- Bifurcation treats contraries as if they were contradictories.
- Examples:
- “If you know BMWs — either you own one, or you want one.” (You could be indifferent.)
- “We must choose between a good car and a cheap one.” (There may be a middle-range option.)
- “You’re either with us or with the enemy.” (Classic political bifurcation — Bush, 2001.)
- “In life, you either choose family or career. You choose.” (Many people successfully balance both.)
Why Fallacies of Presumption Are So Pervasive
Two psychological phenomena explain why these fallacies are deeply entrenched:
False consensus effect: We tend to believe a claim because we falsely believe it is what other people also believe. This cognitive bias — existing long before social media echo chambers — means we protect beliefs from scrutiny by implicitly assuming everyone shares them.
Conjunction problem (Tversky & Kahneman, 1982): In a famous experiment, subjects were given a profile of “Linda” — 31, single, outspoken, philosophy graduate, deeply concerned with discrimination and social justice. Asked to rank the probability of statements about Linda, 89% of ordinary subjects (and 85% of Stanford grad students) judged “Linda is a feminist bank teller” more probable than “Linda is a bank teller” — even though P-and-Q can never be more probable than P alone.
Why does this matter for bifurcation? The researchers had “rigged” the set of choices so they weren’t genuine alternatives (exclusive and exhaustive). Subjects didn’t even notice that (h) and (f) were in the relation of P-and-Q vs. P. The lesson: when alternatives are presented to you, ask whether they are genuine alternatives — are they truly exclusive and exhaustive? Bifurcation exploits this by presenting contraries as if they were the only options.
Distinguishing Sweeping vs. Hasty Generalization
| Sweeping Generalization | Hasty Generalization | |
|---|---|---|
| Direction | General rule → particular case (blocked by special circumstance) | Particular case → general rule (special case can’t support it) |
| Problem | Applying rule where it doesn’t apply | Creating a rule from an unrepresentative case |
| Also called | Fallacy of accident | Converse accident / secundum quid |
| Memory aid | ”Sweeping” the rule over the blocker | ”Hasty” jump from one example to a universal claim |
Do not confuse with Composition (feature of each individual → claimed feature of the whole group) or Division (feature of the group → claimed feature of each member). Those are fallacies of ambiguity, not presumption.
graph TD A[Fallacies of Presumption] --> B[Sweeping Generalization] A --> C[Hasty Generalization] A --> D[Bifurcation] B --> E[General rule applied to a case with special circumstances] C --> F[Special case used to construct an unwarranted general rule] D --> G[False either-or: contraries treated as contradictories] E --> H[Violates boundary conditions of the rule] F --> H G --> I[Other alternatives exist but are excluded]
(diagram saved)
Cross-Course Connections
InformalFallacies — parent category
CategoricalStatements — contradictories vs. contraries distinction (from categorical logic)
FallaciesOfAmbiguity — Composition and Division are related but distinct (Unit 6)
FallaciesOfEmotionalBias — Unit 9 companion: ad hominem, mob appeal, etc.
FallaciesOfEvadingTheFacts — Unit 9 companion: straw person, begging the question, etc.
Bias — false consensus and social conservatism drive presumptuous reasoning
Key Points for Exam/Study
- All three embed hidden assumptions that have not been proven
- Sweeping: rule → case, blocked by a special circumstance (accident)
- Hasty: special case → rule; sample is too small, biased, or exceptional
- Bifurcation: presents only two options when more exist; confuses contraries with contradictories
- The “just because” test catches hasty generalizations: “just because this one case happened doesn’t mean it’s the norm”
- Bifurcation is easy to spot: claim that only two possibilities exist when you can think of a third
Open Questions
- How do we judge when a sample is large enough to avoid hasty generalization? Is there a principled threshold?