How to use this guide
Each entry gives the work, the document that records it, whose name ended up on it, and the mechanism that put it there. Where a claim travels further than its evidence — in either direction — there is a “how it circulates / as it checks out” table. Click any entry to open its field notes. Every quotation is traced, and the route to each source is marked in the footer.
Why the sort is the mechanism
The obvious sorts all failed. By discipline you get physics, genetics and chemistry, which tells you nothing. By fame you are building a ranking, and a ranking of people who were overlooked is a strange object. By how badly wronged, you are asking this guide to award damages. What actually separates these is the mechanism — and that is the useful sort, because a reader who can tell an authorship rule from a stolen slide from a story that dropped a name already on the page can see it happening rather than only reading about it afterwards.
What this guide will not do
It will not sharpen a true complaint with a false detail, which is the commonest way this genre goes wrong and the fastest way to lose the argument. Two entries here exist mainly to correct the story in the direction that is less flattering to the case: Jocelyn Bell is on the paper, and Hedy Lamarr did not invent Wi-Fi. A props culture that overclaims is not a props culture. The way to honour someone is to say what they actually did.
Five ways a name comes off
Say “she was not given credit” and you have collapsed five different things into one. The work was split, and only one half counted. The credit followed the senior person in the room. The material moved, and the name did not travel with it. The name is on the document, and the story dropped it anyway. And the correction arrived, very late. They are not degrees of the same thing. They leave different evidence, they are caught by different checks, and only some of them involve anybody deciding to do it. Five sections, and a sixth where the form stops.
One · The work was split
Two findings where one half took the prize and the other half made it mean something
Two · Credit ran uphill
Two cases resolved toward the senior man without anyone having to decide
Three · The material moved
Three where slides, a photograph and a method changed hands and the name did not
Four · The story dropped it
Two where the document is correct and the retelling is not — in both directions
Five · The correction, late
Two that were put right, and the length of the gap
One · The work was split
When one half of a finding takes the prize
Both of these turn on a division that looks administrative and is not. A result and what the result means are different pieces of work; so are a prediction and the experiment that decides it. In each case the halves were done by different people, and the prize went to one half. Neither case needs anybody to have acted badly for the outcome to be what it is — which is exactly why naming the mechanism matters more than naming a villain.
Two · Credit ran uphill
When it settles on the senior person by default
No decision is recorded in either of these, and that is the finding. A simultaneous discovery gets attributed to the better-known of the two; a laboratory’s work gets attributed to whoever runs the laboratory. The mechanism is a default, not an act, which makes it the hardest of the five to point at and the easiest to reproduce. It is also the one most directly answered by the ordinary habit this guide is about: saying, out loud, who did which part.
Three · The material moved
When the thing changes hands and the name does not
Slides, a photograph, a preparation method. In all three the physical work left the person who made it and arrived somewhere better resourced, and the attribution did not make the journey. These are the most concrete cases in the guide and the least ambiguous, which is why two of the three eventually produced formal corrections — one of them sixty-six years later, and one within six years by a doctor who simply refused to let it stand.
Four · The story dropped it
When the document is right and the retelling is not
These two exist to keep the guide honest, and they point opposite ways. In one, the record is better than the story people tell about it, and repeating the popular version does the person an injury while claiming to defend her. In the other, the record is real and the legend built on top of it is not, and the legend is the flattering one. A catalogue that only ever corrects in the direction of its own argument is not a catalogue.
Five · The correction, late
When it gets put right, decades on
The two here were corrected, publicly and properly, and both corrections arrived long after they would have changed anybody’s career. They are in the guide because a correction is a measurable thing and the gap is the measurement — the same move No. 62 makes with seven crossings. A fixed case is not a case that did not happen.
Six · Where the form stops
Three the apparatus cannot take
Three cards with no reframe table, dashed, not numbered, and not filed as entries — the way FG 10, FG 17 and FG 18 mark theirs. One is not a credit case at all and running a credit frame over it would do real harm. One is genuinely contested in both directions and this guide will not settle it by picking the side that suits. And one has nothing to correct yet, by definition. A subject that cannot take the apparatus gets a different card, not a smaller one.
Not:
Not a ranking. This is the refusal the whole guide is built around and it is first for that reason. The entries are sorted by mechanism, and the sections are not degrees of severity. Ordering people by how badly they were wronged asks this guide to award damages, which it has no standing to do and no evidence for.
Not an argument that these women were greater than their colleagues. Hahn’s chemistry was real, Wilson’s cytology was real, Lee and Yang predicted something extraordinary, and Watson and Crick built the structure. Nothing here is improved by shrinking somebody else, and a case that needs the other person diminished is a weak case.
Not a claim that every one of these was deliberate. Two of the five mechanisms operate with nobody deciding anything — a default that resolves toward eminence, a category that does not yet exist — and one has a stated political cause that is not sexism at all. Naming a villain where there is a rule leaves the rule running.
Not sharpened. The two entries in section four are there specifically to correct the story in the direction that costs the argument something: Jocelyn Bell is on the paper, and Hedy Lamarr did not invent Wi-Fi. A catalogue that only ever corrects toward its own conclusion is not a catalogue, and a grievance carried on a false detail is dismissed on the detail.
Not a set of exceptional cases. The three cards in section six say what the eleven cannot: this guide holds the cases somebody already recovered, each by an identifiable act of work, and there is no way to estimate the ones nobody has done that for. Eleven is not a sample.
Not about credit, in one place. Henrietta Lacks is not an entry and carries no reframe table, because tissue taken without consent is not an attribution dispute and describing it as one proposes a remedy that addresses none of it. The format difference is the argument, the way FG 10 and FG 18 use it.
The line:
Most guides in this collection arrive at there is no standard star, nest, egg, shark, turtle, way to rest, way of being family. This one arrives at the same shape with a different subject, and it is the finding rather than a flourish: there is no standard way for a name to come off a piece of work. Five mechanisms across eleven cases, and they do not reduce to each other — an authorship rule that admits three people, a default that resolves toward whoever is already known, a camera in somebody else’s building, a retelling that degrades a correct record, and a discipline with no name yet. Which is why there is no single rule to guard against it. You cannot fix a default with a policy about theft, or a missing category with a rule about authorship. What reaches all five is the ordinary, repeatable, slightly awkward habit of saying who did which part, out loud, in front of people, especially when they are not in the room to say it themselves. Zine No. 109 asks a four-year-old for exactly that, in four words.