Study guide
This chapter groups three related skills that all require abstracting away from an argument's specific subject matter to its underlying structure: applying a stated general principle to a new concrete situation, matching one argument's logical pattern to another with completely different content, and naming the argumentative technique or rhetorical move an author (or a respondent) uses. Each skill draws on the flaw catalogue from Chapter 3 and the conditional-logic tools from Chapter 4, applied at a higher level of abstraction.
Applying Principles to New Cases
Principle application questions give a general rule, often phrased as a conditional ('a council member should recuse herself if...') or as a broad normative claim, and ask which specific scenario the rule supports, requires, or forbids. The central skill is reading the principle as precisely as a legal rule: identify exactly which conditions trigger the principle and exactly what result it mandates once triggered, then check the proposed application against those conditions term by term. A frequent trap adds a condition to the principle that was never stated, such as requiring that a conflicted bidder be 'likely to win' before recusal is triggered, when the actual principle triggers on the mere existence of a financial stake, without qualification. Another trap misapplies the principle to the wrong party or the wrong action — for instance, a principle governing when a council member must recuse herself cannot, by itself, support disqualifying a bidding company, since the rule was never about what happens to bids. Some questions run the reverse direction, asking you to identify which principle, if valid, would justify a given argument's reasoning; here the correct principle is the one that, once assumed, converts the stated facts into a valid case for the stated conclusion, in the same way a sufficient assumption does in Chapter 1.
Identifying Parallel Reasoning (Valid Arguments)
Parallel reasoning questions ask you to find, among five unrelated arguments, the one whose logical structure matches the stimulus's structure most closely, regardless of subject matter. The first step is always to abstract the stimulus into a skeleton using letters for its terms, ignoring content entirely: for example, 'every A is B; no B is C; therefore no A is C' describes a valid universal syllogism that could be about ferries and diesel power just as easily as about violins and fragility. The second step is to check each answer choice against that same skeleton, paying close attention to whether each premise is a universal ('all,' 'no'), an existential ('some'), or a conditional, since swapping a 'some' premise in for a universal one, or vice versa, breaks the parallel even when the surface topic and conclusion direction look similar. A frequent trap answer preserves the general topic or feel of the original argument, or reaches a conclusion that sounds similarly confident, while actually using a different logical form, such as illicitly converting a term the original argument never converted. Because these questions are often long, the time-efficient approach is to diagram the stimulus once, rigorously, and then test each choice against the diagram rather than re-reading the stimulus for each choice.
Parallel Flaw Questions
Parallel flaw questions are a specialized version of parallel reasoning in which the shared structure is a specific logical defect rather than a valid pattern, and the task is to find the answer choice built on the identical flaw, from the flaw catalogue in Chapter 3. As with parallel valid reasoning, the key is to abstract the stimulus's flaw into its skeleton form first — for example, recognizing an argument as 'affirms the consequent' (if A then B; B; therefore A) — and then testing each choice against that same skeleton rather than against a vague sense of 'this also seems flawed.' The traps here are especially sharp because several answer choices will contain some flaw, just not the same one: one choice might deny the antecedent instead of affirming the consequent, another might commit a valid modus ponens or modus tollens inference with no flaw at all, and only one will replicate the exact defect. A separate common parallel-flaw pattern involves comparing percentages in the stimulus and concluding something about absolute numbers (or vice versa) without base-size information; correctly parallel answers must run the conversion in the identical direction (percentage-to-number, not number-to-percentage) that the original stimulus did. Precision in stating the flaw's direction, not just its category, is what separates the correct choice from its closest competitor.
Method of Argument / Argumentative Technique
Method of argument questions ask how an argument achieves its persuasive goal — what rhetorical or logical move it makes — rather than whether the argument is good or how to complete it. Common correct-answer techniques include: citing a comparable case or counterexample where the predicted effect did not occur (or the opposite occurred) to rebut a prediction; pointing out that an opponent's evidence is equally consistent with an alternative explanation; conceding a claim for the sake of argument and showing the conclusion still fails even if it is granted; and identifying a self-selection or confound in an opponent's causal evidence. These questions are frequently paired with two-speaker stimuli, where a second speaker responds to a first speaker's argument, and the task is to characterize precisely what the second speaker's rebuttal does — for example, offering an alternative explanation for a correlation the first speaker treated as causal, rather than disputing the underlying data or attacking the first speaker's motives. The most common trap answers mischaracterize the response as more aggressive or more concessive than it actually is: describing a targeted evidentiary rebuttal as 'questioning the opponent's motives,' or describing a full evidentiary challenge as merely 'granting the point but disputing its relevance.' Reading the second speaker's exact words and matching them, clause by clause, against the proposed characterization is the most reliable safeguard.
Key terms
- Principle application
- — A question type asking which specific scenario is supported, required, or forbidden by a stated general rule, based strictly on the rule's stated trigger and result.
- Parallel reasoning
- — A question type asking which answer choice's argument shares the same logical structure as the stimulus, independent of subject matter.
- Parallel flaw
- — A parallel reasoning variant in which the shared structure to be matched is a specific logical defect rather than a valid pattern.
- Argument skeleton
- — An abstracted, letter-based representation of an argument's logical form, stripped of its specific subject matter, used to compare structures across arguments.
- Method of argument
- — A question type asking how an argument achieves its persuasive aim — e.g., by counterexample, alternative explanation, or concession — rather than asking whether it succeeds.
- Self-selection
- — A confound in which participants choose whether to join a group being studied, so that a pre-existing trait of the volunteers, rather than the studied intervention, may explain an observed difference.
- Two-speaker (dialogue) stimulus
- — A stimulus format in which a second speaker directly responds to a first speaker's argument, tested by method-of-argument or point-at-issue questions.
Exam tips
- On principle application questions, list the principle's trigger condition and its mandated result separately before reading any application scenario, then check the scenario against each in turn.
- Never let a principle-application answer choice add an unstated qualifier (such as likelihood, size, or degree) to the rule; if the principle doesn't mention it, the answer can't require it either.
- Diagram the stimulus's argument structure using letters (A, B, C) before reading any parallel reasoning answer choice, and test each choice against that same diagram rather than its surface topic.
- In parallel flaw questions, expect at least one distractor to contain a different flaw and at least one to contain no flaw at all (a valid modus ponens or modus tollens) — don't select an answer just because it 'also seems weak.'
- For method-of-argument and two-speaker questions, quote the responding speaker's exact function to yourself ('offers an alternative explanation,' 'concedes but shows irrelevance') before scanning the choices.
- Percentage-to-number parallel flaws must be matched in the same direction as the original; a choice that runs number-to-percentage is a trap, not a match.