Week 11: What prose cannot force you to say

Why a written justification can hide exactly the weakness a live question exposes in seconds, and what the Court is actually testing for

17 May 2027

Open the slides

Week 7 put you across the table from one person, in private, with a pressure log you could revise afterward if you wanted to think it over. Today’s Court is public, panel-based, unscripted, and marked the moment you speak. The difference is not cosmetic, and today is about why it matters.

The paper already said this could be tested

A written disclosure exists so that a challenge can be aimed at something specific. This is the second half of the same duty the Federal Court places on expert reports: they must state the assumptions and material facts an opinion rests on, so that any challenge to it can be aimed at something specific rather than at a guess. Disclosure and challenge are one mechanism, not two — a report that hides its assumptions cannot be usefully cross-examined, and an examination that ignores the report is not testing the estimate, it is testing improvisation.

What a sentence can survive that an answer cannot

A written justification only has to survive being read. It can be dense, qualify itself twice, and lean on a phrase doing more work than the writer noticed — a reader moving at their own pace will often let all three pass. A live question does not move at the reader’s pace. Asked to explain, in the next breath, exactly why one number rather than another, the same assumption either produces a specific answer or it does not. Vagueness that reads as thoroughness on the page reads as evasion out loud, within about two seconds, to everyone in the room at once.

This is not a claim that written work is less honest. It is a claim that prose and live examination are different tests, aimed at different failure modes: prose exposes whether an argument holds together end to end; live questioning exposes whether the person who wrote it can defend any single joint in it, isolated, on demand, without the rest of the document to lean on.

The question the Court is actually asking

You already know accept, reject, and widen — week 7 trained the vocabulary, and using it again today teaches you nothing new. What is genuinely being tested this week is something the vocabulary doesn’t capture: whether the assumption survives being said out loud to someone who is allowed to keep pushing. A number can be perfectly defensible on paper and still belong to someone who cannot explain, live, why it is defensible — and a number that turns out to be wrong can still belong to someone whose live reasoning about it was excellent throughout. The Court is built to tell those two people apart, which a written submission on its own cannot do.

Outline