Offense and Defense: The Difference Between Beating an Argument and Winning a Round

Here is the round that produces the same conversation with a coach every season.

The student's flow looks excellent. Every affirmative contention has a response written next to it, the responses are specific, the evidence was good, and the student sits down certain they won. The judge votes affirmative. The ballot says something like: the negative made me doubt the affirmative case, but never gave me anything to vote for.

Nothing on that flow was wrong. The problem is that all of it was defense.

Offense is a reason you win. It is anything that supports your side of the resolution, anything a judge could point at and say, this is why this debater won the round.

Defense is a reason your opponent is wrong. It shows that something they said does not hold up. It says nothing about why you should win.

Keep it in that short form, because that is the version you can actually use while a round is happening. Offense is for me. Defense is against them.

The gap between the two decides most rounds, because of what a judge has to do at the end of one. They have to write down a reason somebody won. Offense hands them that sentence. Defense only empties out the other side of the page, and an empty page on the other side is enough only when your own page is not empty too.

That is what happened to the student in the round above. Every response on their flow said the affirmative is wrong. Nothing on it said the negative is right.

In the last minute of prep before a rebuttal, ask one question: if the judge believed only the arguments I am currently winning, what sentence do they write on the ballot?

If the answer is "the affirmative did not prove their case," you have an argument for why the round is close, which is not the same as an argument for why you won it.

If the answer is "the negative showed that treating colonization as an obligation makes the risk the affirmative cares about worse," you have offense, and the rest of the speech is about protecting it.

Run that test out loud. Students who run it silently tend to give themselves credit for a sentence they never actually said in a speech.

Not all defense is the same size, and knowing which kind you have tells you how much it can carry.

Non-uniqueness says the thing they are worried about is already happening, or already prevented, regardless of which way the judge votes. Their impact is real and the ballot does not change it.

A link takedown says their advocacy does not actually cause the thing they claim it causes. This is usually the strongest defense available, because an impact with no link is worth nothing no matter how large it is.

An internal link takedown says the chain holds at the start and the end but breaks somewhere in the middle. These are the responses good debaters find and novices skip, because finding them requires reading the other side's evidence closely enough to see the step it skipped.

Impact mitigation says the harm is real but smaller, slower, or less likely than claimed. On its own it rarely wins anything. Its job is to shrink their impact so that your offense is comparatively bigger.

Debaters split these into terminal defense, which takes an argument to zero, and mitigatory defense, which shaves it. Terminal defense is rare and worth naming out loud when you have it, because judges discount claims of it by default and will not give you the zero unless you argue for it.

A turn is a response that takes the other side's argument and makes it a reason to vote for you. Mixing up the two kinds causes one of the worst accidents in debate.

A link turn accepts that their impact would be bad and argues their advocacy causes the opposite of what they claim. They say settlement reduces existential risk; you argue settlement increases it, because the technologies and the resource competition it requires raise near-term risks faster than any far-future benefit arrives.

An impact turn accepts their link and argues the thing they are calling good is bad. They say settlement accelerates technological development; you argue accelerated development of exactly those technologies is dangerous, and that the acceleration is the harm.

Either one converts a response into a reason to vote for you. Running both at once does not double your offense. It cancels it.

That is the double turn, and it is worth being able to see coming. If you argue that their plan causes less of X, and also that X is good, you have just argued that their plan is bad at producing something good, which is an argument for their side. Good opponents will notice and spend the next speech thanking you. Before running a turn, say the combination out loud as one sentence and check which direction it points.

Your own case, extended with its warrant and its impact intact. Extension means restating the argument with the reason attached, not saying the words "extend contention one."

Turns on their arguments, as above.

Framework offense, meaning the criterion itself gives the judge a reason to prefer your side before the contention debate is even resolved.

And independent positions, like a counterplan, a kritik, or a theory argument at higher levels, which bring their own offense into the round rather than borrowing from the case debate.

Defense is cheap. A mitigation response can be one sentence and still sound strong. Offense costs several, because it has to carry a warrant and an impact with it, and a student giving a four-minute speech against a seven-minute one feels every sentence.

So the incentive under time pressure runs toward the cheap thing, and a rebuttal turns into a list of reasons the other side is not that impressive. The fix is a rule you can follow while panicking: no rebuttal ends without extending at least one complete piece of offense, with its warrant and its impact said out loud. Cut a defensive response to make room if you have to.

Pure defense does win rounds sometimes, and pretending otherwise would leave out the situation where a judge decides a debate in which neither side has offense.

That default is called presumption. Many judges presume negative in Lincoln-Douglas, on the reasoning that the affirmative is the one asking for a change and has not justified it. Some presume for whoever is winning the framework. Some resolve it differently again, and it is one of the more genuinely contested questions in judging philosophy, which is why a judge's published paradigm is worth reading before the round.

Winning on presumption means winning because nothing else was left. It is a place to land, not a plan.

Most competitive rounds end with offense on both sides, and then the round is decided by comparison rather than by who had more arguments.

Magnitude asks how much is at stake. Probability asks how likely the chain actually is, and this is where all that defense finally earns its keep, because mitigating their probability is what makes your smaller impact the larger expected one. Timeframe asks what happens first, which matters more than novices expect, since an impact that arrives in a century can be prevented by the one that arrives in a decade.

Then the criterion link, which in Lincoln-Douglas outranks the rest. An impact the criterion does not measure is an impact the judge has been told not to count.

Weighing is done in sentences that contain the word "even if." "Even if they win their contention is true, our impact comes first, because theirs depends on a settlement program that does not exist for decades and ours is running now." Judges reward that sentence more than any single new response, because it does their job for them.

The line-by-line refutation drill builds the habit of answering the actual argument in front of you. The extension tracker shows whether your offense is still alive by the last speech, which is the question this all comes down to. If you want a coach to work through a round with you, that is what a session is for.

Also in this series: how one argument is built, the round clock, and case construction.