Decision Autopsy
An individual close reading of one real, publicly documented decision — what its paper trail can and cannot show about where the decision actually formed, argued with evidence rather than assumed.
Due: 29 March 2027 at 12:00 pm AEDT · Weight: 20%
The brief#
Take a real decision you were not part of, and read its paper trail the way Weeks 1–5 taught you to: as evidence, not as a story someone already told you.
Pick a decision with a documented trail: minutes, an agenda, an announcement, a press release, a public consultation record. A council rezoning, a university policy change, a company’s product shutdown, a sporting body’s ruling — anything with dates you can put in order and documents you can link to. The point is not to guess at motives you cannot evidence. It is to show, from the record itself, what it does and does not support — about where the real decision happened relative to the meeting that announced it, and about how confident the record itself lets you be.
A strong autopsy does not need to find a seam. If your timeline shows a process that was genuinely open until the recorded meeting, or shows evidence that is currently indeterminate either way, say so and argue it — that is not a weaker answer than finding a pre-decided point, provided you show the same rigour: the alternative explanations you considered, and why the record supports the conclusion you reached rather than another one.
What you submit#
A written autopsy, 1,200–1,600 words, structured as:
- The decision, stated as a single sentence, with a link to where it was announced.
- The timeline, every document you can date, in order, each with a one-line note on what it shows.
- The seam, or why there isn’t one — one of three conclusions, each
acceptable if the argument is evidenced:
- A seam: the specific point in the timeline where the outcome looks settled before the forum that supposedly settled it, the evidence that supports it, and at least one alternative explanation you considered and could rule out, with the reason why.
- A genuinely open process: a reasoned case, from the timeline, that no such point exists.
- Indeterminate: a reasoned case that the record cannot currently decide between a seam and an open process — naming the specific evidence that is missing (the thing that, if it existed, would settle it), a specific alternative explanation you could not rule out with what the record actually contains, and what would change your judgement if that missing evidence turned up. Being unable to rule an alternative out is what makes a conclusion indeterminate; ruling one out belongs to the first conclusion above, not this one.
- What you cannot show, the gap between what the documents prove and what you suspect. Naming this gap honestly is part of the mark.
Cite every document you use. A claim with no citation does not count as evidence, even if it is true.
The spec
The brief poses the problem and leaves room for a response. The spec is the fixed part: what a marker checks. Some lines can be checked mechanically; the rest call for judgement.
- names a specific, real, publicly documented decision you did not make
- reconstructs its timeline from primary sources you can cite, not summaries of them
- reaches a specific, evidenced conclusion about whether the record supports a pre-decided outcome, an open process, or is genuinely indeterminate — argued from the timeline, not asserted
- considers at least one alternative explanation for whatever seam or gap you identify, and says why your reading beats it, or admits it doesn't
- distinguishes what you can show from what you are inferring
How it is marked
| Criterion | Weight |
|---|---|
| Quality of the evidence read | 40% |
| Clarity of the account | 30% |
| Judgement about what mattered | 30% |