News and Views

A thousand words tracing a circulating claim back to its primary source, and reporting the gap between what the source says and what everyone repeats.

Due: 9 April 2027 · Weight: 25%

Journals run a short commentary section on recent papers — News and Views at Nature, Dispatches at Current Biology. A piece runs to about a thousand words, takes a position rather than summarising, and is written for someone competent but outside the immediate field.

You are writing one of those, with an investigative constraint: take a claim that circulates, find where it came from, and report the difference.

The exemplar#

You met it in week 2. The mantis shrimp is described almost everywhere as seeing millions more colours than we do, on the reasonable grounds that it has twelve photoreceptor types against our three. When Thoen and colleagues tested wavelength discrimination in 2014, it came out roughly ten times worse than a goldfish.

The popular claim does not overstate the finding. It inverts it. That is the shape of thing worth a thousand words.

Choosing a claim#

Good candidates have a specific primary source, a specific circulating version, and a gap you can characterise precisely. Several are already on the reading list, which is not a reason to avoid them — arriving at your own account of the Dunning-Kruger exchange is real work.

What does not work is a claim with no identifiable source, or one where the gap turns out to be that the popular version is simply less detailed. Less detailed is not wrong, and a piece that spends a thousand words discovering that journalists simplify has found nothing.

The most interesting outcome available to you is finding that the claim survives contact with its source. If you investigate honestly and the popular version holds up, write that. It will mark well, and it is rarer than you would think.

What to submit#

One document. No figure is required; include one if it does work that prose cannot.

Bring your chosen claim to Rosalind or me by the end of week 5. Most of the failures on this task are choice failures rather than writing failures, and they are cheap to prevent and expensive to fix.

The rules that apply to this task#

Extensions, late penalties, academic integrity and what you may use a language model for are on the policies page, which governs all four assessments. Nothing on this page overrides it.