What exactly counts as "the issue"?
The issue is the question of law the court answered in order to dispose of the appeal. Three features distinguish it from everything else in the opinion. It is answerable yes or no (or "affirmed/reversed"). It is necessary to the outcome — if the court could have reached the same result without answering it, you are looking at dicta, not the issue. And it is tethered to facts. "Was the defendant negligent?" is not an issue statement; it is a topic.
A usable issue has a rule component and a fact component fused together. Compare two versions of Palsgraf v. Long Island Railroad Co., 248 N.Y. 339, 162 N.E. 99 (1928). Weak: "Was the railroad liable?" Strong: "Does a railroad whose employees negligently dislodge a package from a boarding passenger owe a duty of care to a plaintiff standing far down the platform, outside the range of apprehended harm?" The second version tells you what the case is good for. The first tells you nothing you could use on an exam.
Where in the opinion should you look first?
Most modern appellate opinions announce the issue in the first two paragraphs or at the start of the analysis section. Scan for stock phrasing and mark it: "The question presented is whether," "We granted certiorari to decide," "This appeal requires us to determine," "The sole issue on appeal is," "We must decide whether." Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803), is the extreme case — Marshall lists three numbered questions and then answers them in order. When the court does that work for you, quote it and move on.
Also hunt for negative signals, which are just as informative: "We need not decide whether," "We assume without deciding," "Because we resolve the case on other grounds." Those sentences fence off what is not the issue, and professors love to ask what the court left open.
If nothing is labeled, read the last paragraph. The disposition — affirmed, reversed, vacated and remanded, certified question answered — tells you what the court was actually deciding, and the reasoning immediately preceding it contains the issue.
How do you reverse-engineer the issue from the holding?
Older opinions, and many state-court opinions, never state a question. Pierson v. Post, 3 Cai. R. 175 (N.Y. Sup. Ct. 1805), just argues about Justinian and pursues its way to a result. Use the inversion method: write the holding as a declarative sentence first, then flip it into a question.
Take MacPherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050 (1916). Holding: a manufacturer of a finished product that is reasonably certain to place life and limb in peril if negligently made owes a duty of care to a remote purchaser despite the absence of privity. Invert: "Does an automobile manufacturer owe a duty of care to a consumer who bought the car from a dealer rather than from the manufacturer?" That is your issue, and it took ten seconds once the holding was written down.
The inversion also audits your holding. If flipping the holding produces a question the court plainly did not answer, your holding is too broad and you have swept in dicta.
How does procedural posture narrow the issue?
The posture supplies the verb of the issue statement, and getting it wrong is one of the most common briefing errors. An appeal from dismissal under Fed. R. Civ. P. 12(b)(6) presents whether the complaint's well-pleaded allegations state a plausible claim — the court assumes the facts are true and decides nothing about who wins. An appeal from summary judgment under Fed. R. Civ. P. 56(a) presents whether a genuine dispute of material fact exists. An appeal from denial of a Rule 50 motion presents whether legally sufficient evidence supported the verdict. A discovery sanction or evidentiary ruling usually presents whether the trial court abused its discretion.
So the issue in a 12(b)(6) case is never "was the defendant liable." It is "do these allegations, taken as true, state a claim." Write the posture into the question: "On review of a Rule 12(b)(6) dismissal, does an allegation that supervisors knowingly adopted a detention policy because of race state a plausible claim of purposeful discrimination against those supervisors?" That framing keeps you from over-reading the case as a merits ruling.
How narrow should the issue be?
Include every fact the court actually relied on, and no others. The test: strike a fact from your issue statement and ask whether the outcome would still obviously follow. If removing "outside the range of apprehended harm" from your Palsgraf issue changes the answer, the fact is legally significant and stays. If removing "the fireworks were wrapped in newspaper" changes nothing, cut it.
Too broad and the case becomes a slogan you cannot apply to new facts. Too narrow and you have written the fact section again. A reliable length is one sentence of roughly 25 to 40 words, in the form "Whether [rule category] applies when [two or three operative facts]."
Where the majority and dissent disagree about the level of generality, that disagreement is the exam question. Read the dissent's framing of the issue against the majority's; the gap between them is usually the doctrinal fight your professor cares about.
What do you do when a case has more than one issue?
Number them and pair each with its own holding and its own reasoning. Never merge two issues into a compound question. International Shoe Co. v. Washington, 326 U.S. 310 (1945), decides both whether the corporation's activities in Washington supported personal jurisdiction consistent with due process and whether service of process on a salesman plus mailed notice sufficed. Hawkins v. McGee, 84 N.H. 114, 146 A. 641 (1929), decides both whether the doctor's statement about a "hundred per cent perfect hand" could be found to be a contractual promise rather than opinion, and what measure of damages applies.
Then decide which issue you will be called on about. That is a different question from which issues the court decided. Casebook editors cut opinions to a single teaching point, so check three things: the chapter and subheading the case appears under, the note questions that follow it, and the syllabus line for that day. The same case carries different issues in different courses — Erie Railroad Co. v. Tompkins, 304 U.S. 64 (1938), is a vertical choice-of-law case in Civil Procedure and barely a tort case at all.
In your brief, put a pin cite next to each issue pointing to the page where the court frames it, e.g., Int'l Shoe Co. v. Washington, 326 U.S. 310, 316 (1945). Bluebook Rule 3.2 governs the pincite; use "Id. at 320" for the immediately following reference. That habit costs nothing and saves you when you are cold-called and need the page in three seconds.
What mistakes should you check for before class?
Run four quick audits on the issue you wrote. First, is it answerable yes or no? If it starts with "how" or "what," rewrite it. Second, does it name at least two facts from this case? If not, you wrote a topic. Third, does the answer to your question actually dispose of the appeal? If the court could have affirmed either way, you found dicta. Fourth, does your holding answer your issue word for word? Mismatched issue-and-holding pairs are the single clearest sign that the reader did not understand the case.
- Confusing the issue with the parties' contentions — advocates frame questions to win; courts often reformulate them.
- Stating the issue at the level of an entire doctrine ("whether the contract was enforceable") instead of the disputed element.
- Ignoring the standard of review, which converts a merits question into a deference question.
- Treating an alternative holding as dicta, or treating a stray aside as a second issue.