A Lincoln-Douglas case is a structured 6-minute affirmative speech (or 7-minute negative speech) written before the round, built around a value premise, a value criterion, and two to three contentions. Both sides bring pre-written cases to the round. The rest of the round attacks, defends, and weighs them.
Most novice and intermediate LD debaters lose rounds not because they argue poorly in the moment but because their case has structural problems before they walk in. A criterion that does not actually evaluate the value. Contentions that do not link to the criterion. Vague terms that get destroyed in cross-examination. Once the case is broken at the structure level, no amount of clever rebuttal saves it.
This guide walks through how to write one from scratch — what each component does, how the components connect, and how to avoid the case-writing mistakes that consistently sink rounds. There is a flow sheet template you can copy, an annotated affirmative and negative case on a sample resolution, and a section on common errors with the specific fix for each.
For broader format context — speech times, round structure, judging paradigms — see the Lincoln-Douglas complete format guide. This post assumes you know the round structure and focuses entirely on how to write the case.
Every LD case is built from four standard components. They must connect in a specific way: contentions support the criterion, the criterion measures whether the value is upheld, and the value tells the judge what matters most in evaluating the round. Break the chain anywhere and the case collapses.
The overarching moral or political concept the case asks the judge to prioritise. The value is typically a single word or short phrase: justice, morality, governmental legitimacy, individual liberty, human dignity, security, equality. The affirmative tells the judge "this is what matters most in evaluating this resolution," and the negative typically either accepts the value, proposes a competing one, or argues that the affirmative's chosen value is the wrong frame for this resolution.
Strong values are broad enough to be plausibly applicable to the resolution but specific enough to actually do work. "Goodness" is too broad — every case in every round can claim to value goodness. "Compensatory justice in tort law" is too narrow — it only applies to specific kinds of resolutions. "Justice" or "morality" sit at the right level of generality for most LD topics.
The mechanism by which the judge measures whether the value is being upheld. If your value is justice, the criterion answers the question "what counts as just?" Common criteria include maximising utility (the utilitarian framework), respecting Kant's categorical imperative, protecting natural rights, applying Rawls's veil of ignorance, upholding social contract theory, or maintaining the conditions for autonomous agency.
The criterion is the point at which an LD case commits to a specific ethical framework. This commitment matters because every contention has to be evaluated through the criterion. If your criterion is "maximising utility" and your contention is about the inherent wrongness of an action regardless of consequences, the contention does not link — utilitarianism cannot evaluate deontological claims. The criterion controls what kinds of arguments your case can use.
The substantive arguments. Each contention makes a claim about the world and links that claim back to the criterion. Two contentions is standard for a 6-minute AC; three is the maximum. Beyond three, you do not have time to warrant any of them sufficiently — each contention needs roughly 1:30 to 2:00 of speech time including evidence, and the framework intro plus closing weighing eat up another 1:15 to 1:30.
Each contention typically has a tag (your one-sentence summary), one or two sub-points that develop the argument, and at least one card per sub-point. The tag is what your opponent will write on their flow — it should be sharp, memorable, and directly link to your criterion.
The support for each sub-point. Cards are short excerpts from philosophical texts, academic articles, expert statements, or empirical research that back up the contention's claims. Each card has a tag (your one-sentence summary of the card's claim), the author's name and qualifications, the year published, and the cited passage itself. Cards in LD lean heavily on philosophy — Kant, Mill, Rawls, Hobbes, Locke, Nussbaum, Singer — because the value/criterion debate is where the philosophical framework actually does work.
Flowing is the most important note-taking skill in Lincoln-Douglas debate. A flow sheet is a structured page (or set of pages) where you track every argument across every speech, organising arguments vertically and speeches horizontally. The standard LD flow uses five columns — one for each speech.
| Argument | AC | NC | 1AR | NR | 2AR |
|---|---|---|---|---|---|
| Value: Justice | Aff: justice as fairness | Neg V: morality | xtnd, justice = morality | turn: morality ≠ fairness | weigh: justice controls |
| Criterion: Util | VC: maximising utility | VC: cat imperative | util inclusive of CI | cross-app: util fails on rights | util controls weighing |
| C1 (Aff) | Animals suffer → moral consid | resp: suffering ≠ rights | xtnd C1, suff. → moral status | drops C1! | EXT: drop, C1 = round |
| C2 (Aff) | Animals show preference | resp: preferences ≠ moral wt | xtnd: preference = autonomy | turn: prefer harms util calc | de-link turn; xtnd |
| NC (Neg) | — | Moral status = rationality | animals show ration → de-link | xtnd ration ≠ moral status | ration alone fails (CX concession) |
Each row tracks one argument from where it was introduced through how it is addressed in every subsequent speech. The notation conventions vary by debater, but a few are standard: xtnd means "extend" (carry the argument forward without modification), resp means "response," turn means "this argument actually helps my side," drop means the opponent did not address it (a major flag — dropped arguments are usually conceded), de-link means breaking the connection between a claim and the criterion, and x-app or cross-app means an argument applies across multiple flow rows.
The flow sheet does double duty. During the round, it tells you what you have left to respond to — anything you do not cross off is something the other side can extend in their rebuttal. After the round, it shows you the structure of how you won or lost — almost every LD loss can be traced to a specific row on the flow where an argument was either dropped or insufficiently warranted.
Beyond flowing your opponent, you also flow yourself. Before the round, lay out your own case on the AC or NC column — every contention, every sub-point, every card tag. This serves two purposes: it forces you to see whether your case actually has tight structure (if you cannot fit a contention into one cell, the contention is probably bloated), and it gives you a template against which you can flow your opponent's responses without losing your own structure.
Step by step. Each step is a separate decision; getting any of them wrong forces you back to an earlier step.
Every word in an LD resolution is contested ground. Start by listing every defined term and every implicit term. For "Resolved: Justice requires the recognition of animal rights," the contested terms include "justice" (which conception?), "requires" (necessary condition vs sufficient condition?), "recognition" (legal recognition? moral recognition?), "animal rights" (which animals? which rights?). Whichever side controls the definition of each term controls the ground of the debate.
The value should be the moral concept most directly implicated by the resolution. For a justice resolution, the obvious value is justice — picking anything else creates extra work. For a "morally permissible" resolution, the obvious value is morality. For "ought to" resolutions, depending on the topic, the value might be justice, morality, governmental legitimacy, or autonomy.
The temptation as a novice is to pick a clever or unusual value to seem sophisticated. Resist this. Judges reward clear framework debates, not creative ones. Pick the obvious value and spend your sophistication budget on the criterion.
The criterion is where most of the case-writing skill lives. The criterion has to do two things at once: actually measure your value (so the framework is coherent), and provide enough leverage to support your contentions (so the case can win on framework).
For a justice value, common criteria include maximising utility, protecting natural rights, applying the difference principle (Rawls), upholding the social contract, ensuring moral considerability, and respecting Kant's categorical imperative. Each has implications for what kinds of arguments link.
Pick the criterion that best supports your contentions, not the one you find philosophically most compelling. The case is a tool to win this round, not a statement of your personal ethical views.
Write down every argument that could support your side of the resolution. Twenty-plus arguments is reasonable at this stage. Do not edit, do not filter — just list. Each argument should be expressible in one sentence.
Then group the arguments by topic. You will usually find that they cluster into 3 to 5 themes. Those themes are the candidates for your contentions.
From the themes, pick the 2 (or maximum 3) that have the strongest evidence and link most clearly to your criterion. The link to the criterion is the test — if a contention is brilliant but does not connect to your criterion, drop it.
For each contention, write a sharp tag (one sentence, ideally under 15 words). Then break the contention into 2 sub-points that develop the argument. Each sub-point gets at least one card.
For each sub-point, find a card that supports it. Cards in LD are usually 30 to 100 words of cited text. Format: author qualifications, year, then the passage. Trim aggressively — every word of evidence is a word you cannot spend on argument.
The framework intro is the first 30 to 45 seconds of your case. It does three things: defines key terms in the resolution, states your value, and states and explains your criterion. Many debaters under-spend on this section — but the framework is where the round is won or lost more often than not.
Resolution: "Resolved: Justice requires the recognition of animal rights."
Value: Justice.
Criterion: Maximising moral considerability — the principle that any being capable of being morally affected by an action is owed moral weight in proportion to its capacity for being affected.
Why this framework: Justice requires that we give moral weight to all beings affected by our choices. The criterion of moral considerability translates that into a measurable standard: if a being can suffer, can have preferences violated, or can be harmed, justice requires we count that harm.
Tag: Animals are capable of suffering and therefore are morally considerable under any consistent application of justice.
Sub-point A: Sentient animals demonstrably experience pain, fear, and distress (Singer 1975; contemporary affective neuroscience research). The capacity for suffering is the standard the criterion uses.
Sub-point B: Any framework that grants moral weight on the basis of suffering must do so consistently — drawing the line at species membership is arbitrary in the same way drawing it at race or sex would be (Singer's principle of equal consideration of interests).
Tag: Recognising animal rights is the only stable way to enforce moral considerability — voluntary moral consideration fails as a matter of historical record.
Sub-point A: Without legal recognition, the burden of considering animal interests falls entirely on individual moral judgement, which has historically been insufficient (factory farming statistics, conservation failures absent regulatory protection).
Sub-point B: Legal recognition of rights is the standard mechanism by which societies translate moral considerability into enforceable protection — extending this to animals is a logical continuation of the same principle used for vulnerable human populations.
If my opponent argues that animal rights are too costly to recognise, the cost is borne by the moral patients in question — animals — not by the framework of justice. If they argue that rights require rationality, the standard is moral considerability, not rationality, and they have not contested the value or criterion. Vote affirmative.
The Negative Constructive is 7 minutes — one minute longer than the AC. The negative has two structural choices the affirmative does not.
A substantive negative accepts the affirmative's framework (or proposes a similar one) and argues that the affirmative's contentions do not actually support the resolution under that framework. A counter-framework negative attacks the affirmative's value or criterion directly and argues from a different ethical lens.
Substantive negatives are safer — you fight on shared ground and win on the strength of your contentions. Counter-frameworks are higher-variance — if the judge buys your framework you can win in a landslide, but if they do not you have spent half your case on framework debate while your opponent extends contentions unanswered.
Roughly 4 minutes of negative case (own framework, own contentions) plus 3 minutes of direct affirmative refutation is standard. Some debaters go heavier on refutation (5 minutes of NC + 2 minutes of refutation, "frontline" model) when the AC has well-known responses already prepared. Others go heavier on case construction when the AC is unusual and they need to build out the framework debate.
Mirror the AC. Open with framework — your value, your criterion, why it controls — and then 2 contentions linked to your criterion. The same writing principles apply: tight tags, two sub-points per contention, one card per sub-point.
If you are running a counter-framework negative, your contentions should establish that the criterion you propose better evaluates the resolution than the affirmative's criterion does — both internally (your criterion does its own job well) and comparatively (your criterion accounts for considerations the affirmative's misses).
The link between value and criterion is the most under-appreciated piece of LD case construction. Some pairings are intuitive and well-tested; others sound clever but actually break the logical chain.
| Value | Common criteria | Best resolution types |
|---|---|---|
| Justice | Maximising utility · Difference principle (Rawls) · Natural rights · Moral considerability | Distributive resolutions, rights claims, criminal justice topics |
| Morality | Categorical imperative · Maximising utility · Virtue ethics · Care ethics | "Morally permissible" or "ought to" resolutions on individual ethics |
| Governmental legitimacy | Social contract · Consent of the governed · Protecting natural rights | Policy-style resolutions about government action |
| Individual liberty | Mill's harm principle · Negative liberty · Autonomy | Resolutions about individual rights vs collective interests |
| Human dignity | Categorical imperative · Capabilities approach (Nussbaum) · Recognition theory | Resolutions about human-rights violations, treatment of vulnerable groups |
| Security | Hobbesian preservation · Realist state interest · Threshold deontology | National-security and international-relations resolutions |
The pairings above are starting points, not rules. The right framework depends on what your contentions need. If your contentions are about preventing concrete harm, utilitarian criteria support them. If your contentions are about rights violations independent of consequences, deontological criteria are stronger. Picking the wrong framework forces you to defend contentions through a lens that does not actually evaluate them.
The most common framework error. Debater claims "Value: Justice. Criterion: Promoting Democracy." But democracy does not measure justice — democracies can produce unjust outcomes, and just outcomes can come from non-democratic processes. The chain is broken.
Fix: For each value-criterion pairing, write out the sentence "Under this value, X is good if and only if X satisfies this criterion." If that sentence is not literally true, the pairing is broken.
A case has utilitarian criterion but the contentions appeal to inherent wrongness, fairness, or rights independent of outcomes. The contentions might be true and might support the resolution, but they cannot be evaluated through the chosen criterion.
Fix: Every contention must include an explicit link sentence — "This contention is true under [criterion] because..." If you cannot write that sentence in one breath, the link is missing.
Four contentions in a 6-minute speech means roughly 1:00 per contention — not enough to develop any of them properly. The case looks impressive on paper but loses in delivery because each argument is too shallow to survive any response.
Fix: Cut to two contentions. Use the saved time to add a second sub-point or stronger evidence to each.
The case relies on "fairness," "the common good," or "human flourishing" without defining what those mean. The first cross-examination question destroys the case because the debater cannot give a precise definition.
Fix: Define every philosophically loaded term in your framework with a specific citation. "Fairness" should mean "Rawls's first principle of justice" or "equal consideration of interests" — not just "the general idea of fairness."
The case ends with "Therefore I affirm" — but never tells the judge how to weigh affirmative arguments against probable negative arguments. The judge has to do the weighing themselves, which is bad for the affirmative because most judges default to whichever framework was explained more clearly.
Fix: Add 15-20 seconds of comparative weighing at the end of the AC. "If my opponent argues X, here is why my framework still controls. If they argue Y, here is the answer." Do the judge's work for them.
Each card has a tag and a citation, but the cited text is just an assertion — no reasoning for why the assertion is true. Cards in LD have to do more than just establish that someone has said the claim; they have to support the claim with reasoning the judge can evaluate.
Fix: Every card needs a warrant — the part of the cited text that explains why the claim is true. If the card is just an authority's assertion, find a better card or pair it with a second one that provides the reasoning.
The AC is 6 minutes. The NC is 7 minutes. Word counts depend on delivery speed:
Speed in LD varies dramatically by region and judge pool. Traditional circuits (especially novice and JV) reward slower, persuasive delivery; progressive and "circuit LD" national-tournament debaters often speak at speeds approaching Policy debate. Check your judge's paradigm before deciding how much to fit into the case — a 2400-word AC delivered at 400 wpm to a traditional judge is a non-starter regardless of how good the content is.
Time budget within the AC: framework intro around 45 seconds, each contention 1:30 to 2:00, closing weighing 15 to 30 seconds. The 6 minutes have to add up cleanly or the case runs over.
Once written, the case has to be timed, delivered out loud, and refined. Three free tools at DebateClock are useful for this.
The motion practice timer handles prep countdown plus speech timing in one flow. Set the format to LD (6-minute speech default) and the prep time to whatever you want — useful for repeated practice of the same case across sessions to build delivery muscle memory. The flow timer handles structured multi-interval practice sessions; a "Drill" preset (10-minute work blocks with 2-minute breaks) is a good fit for repeating the same case six times in a row with brief notes between attempts. For tracking case timing across multiple practice rounds — checking whether you consistently run over and where — use the practice round logger.
Free LD timer with correct prep pool, all speeches preloaded, two-device sync. No signup.
Open LD timer →