Making it easier for published scientists to change their names

March 18, 2021 1:14 PM

Science said last month: "Today we are pleased to announce a seamless, discreet procedure that authors can follow to change their names in previously published papers across the Science family of journals. We join multiple other publishers that have adopted similar initiatives, including the American Chemical Society, Public Library of Science, Royal Society of Chemistry, and Wiley. Authors may have occasion to change their names for various reasons, but recent outreach by, and on behalf of, transgender scientists has impressed upon us the importance of respecting authors’ privacy and autonomy in correcting the scientific record."
posted by brainwane (23 comments total)

Fuck, this is amazing. I left academia and don't get cited anymore, but having publications in my old name still makes me real sad and squeamish sometimes. I'm sure it's worth even more to people who are still in the middle of an active career.
posted by nebulawindphone at 1:17 PM on March 18, 2021 [21 favorites]

Once upon a time this might have made it difficult for citations to be followed (which would lead to people not using the process since nobody wants to make their papers hard to find) but everyone uses DOIs for finding documents anyway even if the citation format has people's initials and last names in it.
posted by atrazine at 1:40 PM on March 18, 2021 [2 favorites]

Not just done, but done with care and integrity, or so it seems. Great!

(I'll be sending a polite letter to APS and AAS this evening urging them to catch up. It's not unlikely they're already on it and I just haven't been paying attention.)
posted by eotvos at 1:59 PM on March 18, 2021 [1 favorite]

I still haven't changed my publications, but I did ask AIAA and IEEE a while back. AIAA was willing to do it, but IEEE wasn't. Don't remember what APS said. Hopefully that is changing. It would be critical if I ever wanted to change jobs.
posted by Flight Hardware, do not touch at 2:01 PM on March 18, 2021 [1 favorite]

arXiv too.
posted by bdc34 at 2:04 PM on March 18, 2021 [5 favorites]

Flight Hardware, did IEEE give any reasons?
posted by biffa at 2:09 PM on March 18, 2021

Correct citations are the lifeblood of the field. I'd love if Cell, Nature, and Bioinformatics (Elsevier/Oxford) follow suit, and allow name changes without going through the hoop jumping currently required (esp. the requirement to get corresponding authors involved). Important for transgender scientists, and for anyone whose name is not represented correctly.
posted by They sucked his brains out! at 2:09 PM on March 18, 2021

From the arXiv announcement:

arXiv’s new policy aligns with guiding principles recently provided by the Committee on Publication Ethics (COPE), a global organization that aims to integrate ethical practices as a normal part of the publishing culture worldwide.

COPE writes:

We anticipate sharing the initial guidance document—which will likely then continue to evolve over time—within the next few months. We know some journals and publishers are waiting for COPE’s guidance, rather than developing their own policies. For journals and publishers actively developing their own policies or seeking to be proactive, we are happy to put you in contact with members of the working group in the interim, with the aim of providing insight and expertise. More importantly, we know the document cannot come soon enough for trans authors, who are most impacted by the current processes and practices.

So now's a fine time to tell your industry's or field's publications: please pay attention and sign up for COPE's newsletter!
posted by brainwane at 2:23 PM on March 18, 2021 [1 favorite]

This is great!! The requirement to involve all authors that are some places is no good.


This is what APS said last time I asked (Oct. 2020). I will query them again with this info.

Thank you for writing.

At this moment we do not have the capability to "relate" old/new names.
This is part of an author database we shall be working on in the future.

For published manuscripts, those are already published are the recorded
version, and names cannot be updated/changed.

If there is a name change along with an email address change for an
APS Journal account, we can update that account with the new name,
a new username, and email address for that individual. The information
which was previously in that account will remain there.

Please let us know if you should need anything further. Enjoy your day.

posted by lab.beetle at 3:09 PM on March 18, 2021

Awesome! This is also important for people who published as a single person and then changed their names when they got married. I know several people that have had problems due to this.
posted by overhauser at 4:06 PM on March 18, 2021 [9 favorites]

“But we’ll have to update our software to cope!”

Sounds like someone didn’t read Falsehoods Programmers Believe About Names.
posted by Brian Puccio at 4:22 PM on March 18, 2021 [6 favorites]

I love that article, but I think until very recently the biggest issue was not "programmers believing falsehoods" but "everyone treating the print edition as authoritative and expecting the digital version to be a faithful copy." Print is finally dead enough -- dead even for the oldest of emeriti -- that we can now take the digital version as authoritative, the DOI as the crucial part of the bibliography entry, and the print journal with its inked-in permanent author names as a weird ceremonial trophy. But in my field, ten years ago that wouldn't have worked: people still saw the lump of paper in the library stacks as the "real" publication, the data that helped you find it as the "real" metadata, and DOIs as a weird trend that was likely to fail.
posted by nebulawindphone at 4:42 PM on March 18, 2021 [4 favorites]

I guess what I'm saying is, I suspect that for a long time the biggest obstacle was citation practices. I could change my name to whatever I wanted, but people would still feel duty-bound to cite me using the name that appeared in the physical volume in the library, because that physical volume was the real truth and digital copies were mere shortcuts.
posted by nebulawindphone at 4:49 PM on March 18, 2021 [7 favorites]

As with accessibility, designing for inclusivity benefits everyone. I no longer use my middle initials professionally. One of my colleagues changed her last name when she married to be rid of her estranged father’s last name. Another colleague and his partner adopted a new shared last name. All of us have publications kicking around under our old names. Kudos to all of the trans scholars, activists, and allies who saw a problem and did the work to fix it. Out of all the reasons to misidentify trans people harmfully, citation tradition has to be one of the most pointless. Glad to see this change.
posted by grimmelm at 8:21 PM on March 18, 2021 [4 favorites]

What happens if a researcher remembers a paper author by name who has been changed, and tries to search for it using the old name? Does that not tend to happen? I can imagine going to a conference and hearing a talk by someone who later changes their name and not being able to find that researcher's work.

But you do have the other side, where a paper author who has changed their name has most of their work under a different name, which must be a pain when they give talks and then have to explain "My name is $NAME but my papers are mostly authored under the name $DEADNAME."

Is there some sort of author ID system out there, of the random alphanumeric type (or the like)? DOIs seem useful for individual papers but not so much for authors overall. It seems like the assumption that a name is a good static identifier for a person is just bad, especially since we have a centuries-(millenia?-)old tradition of having nearly half the population change their name when they marry.
posted by that girl at 8:44 PM on March 18, 2021

that girl, that’s what the ORCID ID is for - a unique ID per researcher, who controls the attached data, eg name.

Researcher control of ORCID records is one of ORCID’s core principles.

You register for your own ORCID iD, and you always have complete control over your ORCID record, including what information is connected to your iD; what is publicly available, private, or shared with trusted third parties.

But journals, even the ones that expect a ORCID ID, don’t pull the researcher’s name etc. from it on every (any) query. (I assume.) ... If they did it would be closer to normal form, a pun I will Backus away from now, sorry.

The ORCID was developed because of name changes and name collisions (different researchers, same name), as it was explained to me. Nice that itsdesign means it also works for gender transitions.
posted by clew at 9:20 PM on March 18, 2021 [2 favorites]

Ah, good to hear there's at least something, and yes, name collisions are also something that makes searching by name less useful.
posted by that girl at 9:57 PM on March 18, 2021

What happens if a researcher remembers a paper author by name who has been changed, and tries to search for it using the old name?

It’s quite typical to refer to a paper by the name of either the first author or the corresponding author in casual conversation; that is, not as a formal citation practice but in a usual discussion of the literature. Usually we’ll say something like “the Smith PNAS paper from 2015” or “Smith et al. from last year”. I’m worried that I’ll miss a name change in a paper I’m familiar with since I rarely go back to the journal to re-access a paper I know well or have available locally. I’ll be happy to be corrected if I ever mis-name an author though.
posted by mr_roboto at 10:27 PM on March 18, 2021

Glad arxiv is doing some work here too. But they really need to update their search so it works on author records, not on name.

People with highly nonunique names are at a disadvantage; when I click on their name from a paper’s record, arxiv just searches for Johnson, A or Lee, K. And this is a great time to make that change, since the unique author record is being improved to facilitate name changes— just make it the default way to author search.

I’m not sure how to improve searchability for authors who change their names. But generally academics want their work to be findable— and it seems there must be a way to facilitate that even with name changes. Maybe authors could choose to keep an alias list that would show up in searches but with a “did you mean NewName”? (I haven’t and won’t change my name so this only affects me as someone who wants to be able to find the papers of my colleagues, and who wants their work to be findable so they can continue to get the recognitions and citations they deserve—under the name they wish to use).
posted by nat at 1:04 AM on March 19, 2021 [1 favorite]

This isn't just about citations, it'll also help enormously in pulling together an author's publication record. Websites like Scopus or Google Scholar create profiles of authors based on name, affiliation, field etc and generate various metrics about their publishing history. Rightly or wrongly, academic performance is often judged by these metrics, and they certainly influence hiring or promotion committees, although this does vary enormously by field. Ensuring a full a record as possible is used will definitely benefit those who have changed their names, a group often at a disadvantage to their cis male colleagues in those circumstances.
posted by Faff at 1:37 AM on March 19, 2021 [2 favorites]

findable— and it seems there must be a way to facilitate that even with name changes.

We’d have to get used to routing searches for new work through a researcher-ID system, surely? You’d think "what’s new on that nitrogen effect from $NAME?" but you’d have to click on something that took you to the ID, or a disambiguator to pick the $NAME associated with the paper you had been thinking of?

Easier if each journal or PDF hotlinks names to an ID. Maybe that’s the next push.
posted by clew at 1:29 PM on March 19, 2021

All the trans scientists I know have curated their own google scholar profile. It does seem to allow name alternatives. So in that sense, the impact of name changes was already somewhat reduced (to the bean counters, not to anyone who actually wants to be correctly represented) by being able to curate your own publications in pubmed/orcid/google scholar. It just takes more time and work.

Physical review (the APS journal publisher) says this:

Thank you for contacting Physical Review about our name change policy. With the help of other publishers, advocates, and COPE, we've been thinking through a policy, but we have not yet developed an official one. At the moment, we're taking personal requests on a case-by-case basis. Please feel free to contact me directly with any questions.

posted by lab.beetle at 5:55 PM on March 19, 2021 [1 favorite]

clew: Easier if each journal or PDF hotlinks names to an ID. Maybe that’s the next push.

Yes, that's what I was asking for -- except arXiv isn't a journal, it's an open access server. But right now when you click on an author in the arXiv record of a paper, it searches for the person by name, instead of giving you a full list of the person's papers (which it *has* because they have built author records).

I do like the PDF link idea, that is also super important-- I'm not any good at writing bibstyles myself but it'd be awesome if someone wrote a bibstyle for bibtex that hotlinked authors to their inspirehep/arxiv/whatever name record. For extra fanciness they could periodically re-render PDFs on arxiv using this namerecord lookup, so that even citations from years ago could have everybody's current names.
posted by nat at 6:38 PM on March 19, 2021