Boldface and Argument-Structure Questions: Labeling Roles Before Judging Content
Boldface questions ask what a statement does, not whether it is true; role labels answer them faster than content analysis ever will.
Why this matters
Boldface Critical Reasoning questions have a reputation for difficulty that is mostly a reputation for unfamiliarity. They ask about function: is the bolded statement a conclusion, a premise, a concession, or a consideration weighing against the author's position. Candidates struggle because they evaluate whether statements sound reasonable, when the question only cares about the job each statement performs inside the argument.
These items also mark a difficulty checkpoint: they appear more often as candidates perform well, so a dependable method matters for high scorers. The skill they train, structural reading, pays dividends across every Verbal question type, because identifying the author's main conclusion is the first move in strengthen, weaken, and assumption questions as well.
A working method
Read the passage once to find the main conclusion, using signal language and the therefore test: the conclusion is the claim the rest of the passage exists to support. Then assign every other statement a role relative to it: supporting evidence, background, an opposing position, evidence for the opposing position, or a concession the author grants before pushing back.
Only then look at the boldface portions and describe each in your own words, such as fact supporting the author or position the author rejects. Convert answer choices into the same simple vocabulary; the correct answer must match both boldface roles, and most wrong answers describe the first portion correctly and quietly misdescribe the second. Check the second description first, since that is where the exam hides the error.
How to practice this skill
Take ten boldface questions and, before viewing any choices, write a two-word role label for each bolded portion. Then evaluate the choices purely as translations of your labels. This sequencing prevents the answer choices from rewriting your reading, which is the mechanism by which attractive wrong answers succeed on this question type.
Extend the drill to ordinary CR problems: label the role of every sentence in five arguments per session for a week. Boldface competence is a byproduct of general structure-mapping fluency, and candidates who practice role labeling on all arguments report faster and more confident performance on every CR task, not only the boldface family.
A rigorous review protocol
Review by writing the argument as a skeleton: conclusion on top, supports below, opposition and concessions flagged. Place the boldface portions on the skeleton and confirm the credited answer matches their positions. When you missed, identify which role you mislabeled and why, usually a concession mistaken for the author's view or an intermediate conclusion mistaken for the main one.
Watch for the intermediate conclusion pattern specifically: a statement that is supported by one sentence and itself supports another. The exam loves testing whether you can distinguish the final conclusion from a stepping stone. If that distinction caused the miss, drill the therefore test until the direction of support, not the order of sentences, determines your labels.
Applying it in a timed section
Under time, boldface items reward the disciplined sequence and punish re-reading loops. One structural read, two role labels, then choices; if two answers survive, compare only their verbs and their descriptions of the second portion. Cap the item at roughly two minutes, since a stuck boldface question rarely improves with a third reading, and your labels are usually right the first time.
What mastery looks like
Mastery shows when you can hand any argument a skeleton in one read, when concessions and intermediate conclusions no longer masquerade as main conclusions, and when boldface answer choices read like translations to check rather than riddles to decode. The question type then becomes a marker of your strength instead of the exam's.