Reading a draft assertion by assertion

Most editors read a draft the way most readers do: for flow, for pace, for whether the ending lands. That reading matters, but it is a different exercise from checking whether the piece will hold up. For that, it helps to read the draft a second time in a much slower, less forgiving way — one assertion at a time.

An assertion, for this purpose, is any sentence a reader could reasonably treat as a factual claim about a person, an organisation, or an event. Not every sentence qualifies — descriptive colour and background context usually don't — but a surprising number do, more than most drafts account for on a first pass.

Building the assertion list

The method itself is simple, if slow. Go through the draft line by line and, for each qualifying sentence, note in the margin what specifically supports it: a named document, a dated interview, a public record, or a source who has agreed to be identified. Sentences that rely only on "it is understood" or an unnamed source deserve particular attention, since these are the claims most likely to need reinforcement or softening before publication.

Once the list exists, patterns tend to appear quickly. A piece might turn out to lean heavily on a single source for its most serious claims, or have one paragraph where the sourcing thins out noticeably compared to the rest. Seeing this laid out plainly is far more useful than a vague sense that "the middle section feels weaker."

What to do with the gaps

Where support is thin, there are usually three honest options: strengthen the sourcing before publication, adjust the wording to match what is actually known, or cut the assertion if neither is possible in the time available. What should not happen is publishing a claim exactly as written while hoping the weak spot goes unnoticed — reviewers, subjects, and readers tend to find precisely those spots first.

This same pass is also a natural place to check whether every person named unfavourably in the draft was in fact given a genuine chance to respond, and whether the piece reflects that response, or its absence, fairly. A draft that has been through this kind of check reads no differently to the average reader. It simply holds together better under the scrutiny that good investigative work invites.