PHIL 252 Unit 6 — Fallacies of Ambiguity, Meaning, and Representation
Core Argument of This Unit
Most real-world reasoning is informal dialogue, not formal logic. This unit introduces informal fallacies — errors that seem convincing but violate reasoning standards through ambiguity, language misuse, or faulty part-whole reasoning. It also applies critical thinking to visual data, showing how charts and graphs can mislead just as words can.
Key Ideas
What Makes Something a Fallacy: Walton’s five conditions — an argument (or apparent argument) that: (1) falls short of correctness, (2) is used in real dialogue, (3) has a semblance of correctness, (4) poses a serious problem for achieving the dialogue’s goal.
Informal vs. Formal Fallacies:
- Formal fallacy: error in logical structure (e.g., Affirming the Consequent from Unit 3)
- Informal fallacy: error in content, context, or language — often not formally invalid but still fails reasoning standards
The Six Fallacies of Ambiguity:
| Fallacy | What Goes Wrong | Classic Example |
|---|---|---|
| Equivocation | A key word shifts meaning between premises; blocks transitivity | ”Knowledge is power. Power corrupts. ∴ Knowledge corrupts.” (power ≠ power) |
| Amphiboly | Grammatical structure is ambiguous; multiple readings possible | ”Mary and Frieda are visiting doctors.” (Are they doctors, or visiting them?) |
| Accent | Meaning shifts based on where stress or emphasis falls | ”Did you steal the butter?” vs. “Did you steal the butter?” |
| Composition | Properties of parts invalidly attributed to the whole/class | ”Every atom in my arm is invisible, so my arm is invisible.” |
| Division | Properties of the whole/class invalidly attributed to parts/members | ”University graduates earn 70% more on average; Kofi is a graduate, so Kofi earns 70% more.” |
| Hypostatization | An abstract term is treated as if it were a concrete entity | ”The state butts into private affairs.” (as if the state is a person with intentions) |
Composition and Division — the key question: Is the property hereditary? Some properties pass from parts to wholes (a pile of red blocks is red). Others don’t (each musician plays quietly, but together they are loud). Fallacy occurs when non-hereditary properties are treated as hereditary.
Bias: Having a preference, attitude, or point of view. Not all bias is irrational — it becomes a problem when it leads to unfair, inaccurate, or inconsistent reasoning. Critical thinkers identify, scrutinize, and correct for bias. They apply standards universally — to themselves as well as others.
Selection Bias: A systemic error from non-random sampling. When data is collected in a way that reflects interests or preferences rather than the full population, conclusions are distorted (e.g., surveying only ski enthusiasts at a single resort about ski conditions).
Misleading Data Visualizations:
| Problem | Description |
|---|---|
| Duck | Visualization that prioritizes aesthetics over accurate data communication |
| Glass Slipper | Using the wrong chart type for the data (shoehorning) |
| Axis Manipulation | Truncated, inverted, or variable-scale axes that exaggerate differences |
| Bin Width Manipulation | Variable bin widths in histograms shift narrative |
| Proportional Ink Violation | Shaded area doesn’t correspond to actual values |
| Right-Censoring | Omitting cases not yet at endpoint, creating false impression |
Principle of Proportional Ink: When a shaded region represents a number, its area should be directly proportional to that number.
Foundational for Unit 7
- Selection Bias and right-censoring are direct scientific reasoning issues — methodology flaws in experiments
- Composition/Division fallacies appear in causal reasoning (e.g., attributing a group-level cause to individuals)
- Data visualization critical thinking is essential for evaluating scientific graphs and charts
- Equivocation is common in causal language (“causes”, “correlates with”, “is associated with”)
See InformalFallacies, FallaciesOfAmbiguity, Bias, DataVisualization