What is an attack outline, and how is it different from your full outline?
Your full outline is a reference document. It stores rules, case holdings, hypotheticals, and policy arguments so you can look things up. It answers the question "what do I know about consideration?" An attack outline is an operating document. It answers a different question: "a fact pattern is in front of me and the clock is running — what do I do first, and what do I do next?"
The structural difference is sequence. A full outline is organized the way the course was taught. An attack outline is organized the way an exam answer moves. In Torts, your full outline might have a long section on proximate cause with Palsgraf, Wagon Mound, and every intervening-cause case your professor assigned. Your attack outline says: (1) duty owed? (2) breach — run B < PL from Carroll Towing plus any custom, statute, or res ipsa argument; (3) actual cause — but-for, then substantial factor if two sufficient causes; (4) proximate cause — foreseeable plaintiff, foreseeable type of harm, then intervening/superseding; (5) damages; (6) defenses — comparative fault, assumption of risk. Six lines, each of which unpacks into sub-steps.
The second difference is that an attack outline is unusable unless you wrote it. Compression is the learning. A downloaded attack outline is somebody else's compressed memory of a course you did not take.
When should you build it?
Build it after your full outline is substantially complete and before you start heavy practice-exam work — typically the last two to three weeks of the semester. If you make it too early you will be compressing material you do not understand yet, and the outline will be a table of contents rather than a decision tree.
Do not skip the full outline and jump straight to the attack outline. The attack outline is derivative by design. Its brevity only helps you because each line is a compressed pointer to material you already have in your head; if there is nothing behind the pointer, the line is noise. If you are genuinely out of time, write a full outline for the two or three heaviest units and attack outlines for everything else, rather than doing a shallow job across the board.
Expect to revise it after every practice exam. The first version is a hypothesis about how the exam will move. Practice tells you where the hypothesis is wrong.
What goes in each entry?
Each major topic gets one page or less, and that page has four components. First, the trigger: the facts that tell you this topic is in play. Second, the sequence: numbered steps in the order you would write them. Third, the forks: the points where the analysis splits, with the test for each branch. Fourth, the fight: the one or two sub-issues your professor spent class time arguing about, because that is where the points are.
A Contracts entry might look like this. Trigger: any exchange of communications before performance. Sequence: (1) Is this a sale of goods? If yes, UCC Article 2; if no, common law; if hybrid, predominant purpose. (2) Offer — commitment, terms, communication. (3) Acceptance — mirror image at common law versus UCC 2-207 for additional and different terms; note whether both parties are merchants. (4) Consideration — bargained-for exchange, then promissory estoppel as fallback. (5) Defenses to formation — Statute of Frauds, capacity, duress, misrepresentation, unconscionability. (6) Terms — parol evidence rule and interpretation. (7) Breach and excuse. (8) Remedies — expectation, then Hadley foreseeability, certainty, mitigation, and reliance or restitution as alternatives.
Notice what is missing: case names for their own sake, black-letter recitations you already know cold, and background history. Include a case name only when the case is the operative test (International Shoe for minimum contacts, Hanna v. Plumer for the Erie fork, Daimler for general jurisdiction) or when your professor uses it as shorthand in class.
How do you write the forks so they actually function under pressure?
Write forks as conditionals with a destination. "If a Federal Rule of Civil Procedure is on point and answers the question, apply it if valid under the Rules Enabling Act and the Constitution — Hanna. If no federal rule is on point, go to the twin aims of Erie: forum shopping and inequitable administration of the laws." That is usable. "Erie doctrine — Hanna, Byrd, Guaranty Trust" is not, because it makes you rebuild the analysis at the worst possible moment.
Do the same for personal jurisdiction: (1) traditional basis — consent, domicile, or in-state service under Burnham? (2) long-arm statute reach? (3) minimum contacts — general jurisdiction if the defendant is at home under Daimler, otherwise specific: purposeful availment, then relatedness after Ford Motor Co. v. Montana Eighth Judicial District Court; (4) fair play and substantial justice; (5) notice under Mullane. Each number is a place you can stop and write a paragraph.
For Evidence, the fork structure is the whole subject: relevance under 401 and 403, then hearsay — is it an out-of-court statement offered for its truth under 801(c)? If yes, is it excluded from the definition by 801(d)(1) or 801(d)(2)? If not, does an exception in 803, 804, or 807 apply? Then character evidence under 404, 405, and 406, impeachment under 607 through 613, privileges, authentication under 901, and the original document rule under 1002. Write it as that chain, because that is the chain you will run on every objection question.
How should the format change for open-book versus closed-book exams?
This varies by professor, and the format drives the design. For a closed-book exam, the attack outline is a memorization target. Keep it short enough to reproduce from memory on scratch paper in the first five minutes — usually five to eight pages, with mnemonics for element lists. For an open-book exam, it is a navigation tool, and its job is speed of lookup. Add page cross-references to your full outline, and let it run a bit longer, but resist the urge to paste rules back in. The moment it becomes readable prose you have rebuilt the long outline.
Format also changes with question type. For issue-spotter essays, organize by claim and then by element, because that is how you will write. For multiple choice, especially in courses tested MBE-style, organize by distinction: the pairs the test writers exploit, such as larceny versus embezzlement versus false pretenses, or negligent versus intentional infliction of emotional distress. For a policy or short-answer exam, replace some of the element trees with a list of the two or three recurring tensions your professor returns to, plus the strongest argument on each side.
One universal formatting rule: it must be visually scannable. Bold triggers, indent sub-elements, number every step, use one page per topic, and never let a topic run across a page break if you are printing it.
How do you know it works?
Take a past exam from your professor and write the answer using only the attack outline. Every time you have to stop and think "wait, what are the elements of that" or "which test applies here," mark the spot. Those marks are the defects. Fix them by adding a step, not by adding a paragraph of explanation.
Then run the reverse test. Read a fact pattern and, without writing the answer, use the attack outline to produce a one-page issue list in ten minutes. If the outline does not get you to the issues, your triggers are too abstract. Trigger entries should be facts, not doctrines: "defendant is a corporation sued outside its state of incorporation and principal place of business" is a trigger; "general jurisdiction" is not.
Finally, check coverage against the syllabus, not against your full outline. Students routinely under-outline the last two weeks of material because the full outline ran out of steam there — and professors routinely test it.
What are the common failure modes?
The most common is length. An attack outline that runs thirty pages is a full outline with smaller margins, and you will not open it during the exam. If yours is too long, the cut is usually background rules you know cold and case descriptions that duplicate the test they stand for.
The second is topic-order drift, where the attack outline copies the syllabus order instead of the analytical order. Watch for this in Constitutional Law, where courses often teach standing and justiciability late but exams require you to address them first, and in Criminal Law, where defenses are taught as a unit but must be analyzed crime by crime.
The third is omitting defenses and remedies. Students build elegant liability trees and then run out of outline at the moment the exam awards its easiest points. End every claim branch with defenses and damages, even if the entry is three words long.
- Too long to open mid-exam
- Ordered by syllabus rather than by analysis
- Triggers stated as doctrines instead of facts
- No defenses, no remedies, no exceptions to the exception
- Copied from an upperclassman, so nothing sits behind the pointers