Order the argument

Turns the settled thesis and the decisions recorded during research into a section order.

Prerequisite

The working doc docs/research/<slug>.md must contain a ## Thesis section with - [x] Thesis ticked in the stage block. An unapproved thesis means no outline — the order of an argument is derived from its claim, so outlining first means ordering evidence around a claim that may still move.

Most of the outline already exists

Ordering constraints accumulate during research and get recorded as decisions. Harvest them before inventing structure — on this site's SDD post, the ## Decisions block already specified that the 2004 precedent opens and then steps aside, that the quality argument leads over the coverage argument, that the rival framing lands only after the core evidence, and that a polysemous term gets defined early. That is four of roughly seven sections, already settled and already justified.

Read ## Decisions first. Invent only what it does not cover.

Write to the working doc

Append an ## Outline section. One row per section:

## Outline

**Register.** Essay (~7k chars) or grey paper (~50k, full APA reference list). Decide here —
it changes section count more than any other choice.

| #   | Section | Claim it makes | Evidence      | Debt  |
| --- | ------- | -------------- | ------------- | ----- |
| 1   | ...     | one sentence   | which finding | any ⚠ |

Every section needs a claim, not a topic. "Background on SDD" is a topic and will produce filler. "The term was borrowed twice, so it was never specified" is a claim and produces a paragraph that has to earn its place.

Checks before handing off

• Does the order build the thesis sentence? Read the claims column top to bottom. It should compose into the thesis. If it does not, the order is wrong or a section is filler.
• Does the strongest objection get answered? It was named at the thesis gate. Find the section that handles it. If there is not one, add it.
• Is any section carrying a ⚠ that must close before publishing? Mark it. That is the pre-publication checklist, derived rather than remembered.
• Does anything survive that the thesis demoted? Cut it. Demoted findings reappearing in an outline is how a piece turns back into a list of interesting things.

Do not

• Do not outline around the evidence you liked most. Order follows the argument, not the research effort. A finding that took an hour to verify and supports nothing gets cut.
• Do not defer the register decision. "Decide the length while drafting" produces an essay that grows into a grey paper without the structure of one.

Next step

→ blog-write to draft against this outline.