A new essay is up in which I make and explore one specific claim: that when trans people treat narratives of transition or particular metaphors for it as universal norms that capture the reality of what it means to be trans, this constitutes a form of gatekeeping, one we primarily perpetrate against ourselves.

Narratives can be methods for questioning people to consider the possibility of living a trans life, but they can also be walls we build to oppress one another.

https://blog.liliane.io/post/trans-futures-trans-pasts

I'm not working this afternoon, so I'm going to muse out loud for a bit about something I've been confused about for a while.

Feel free to mute me for 24 hours or so if this gets too spammy. 🙂

Question: when we use gendered pronouns for someone, what is it that we're doing?

Anyone know of fun stuff for queer people to do in #philly?

(Day trip ideas, not nightlife)

🚨Report your NSF grant terminations! 🚨

We are starting to collect information on NSF grant terminations to create a shared resource as we have for NIH. The more information we collect, the more we can organize, advocate, and fight back!

https://airtable.com/appGKlSVeXniQZkFC/pag50h1IjrX5MHhCH/form

Please share widely!

#NSF #NIH

How to earn authorship on many authored papers?

The existence of what I jokingly call MAPs (Massively-Authored-Papers) has been a reality in ecology for a few decades now. They are primarily the results of working and synthesis groups. A group of 10-30 people meet up a few times, build consensus, and produce a paper often with 10-30 authors. Love it or hate, this is a major mode of doing science these days. Personally, I love it. I think it has really accelerated the pace at which thinking is shared and limitations are identified and the social hive mind that is science operates. But MAPs raise new questions about paper authorship.

To be clear, I think the basic principles of authorship are pretty widely agreed upon. Fundamentally, you need to make an intellectual contribution. This can come in formation of ideas and methods, data collection, data analysis, writing. Somewhat more controversially funding these efforts may be included (I personally have become increasingly more open to the role of funding as a contribution as I’ve spent more time administering grants and less time collecting data – there’s a surprise). And there is some minimum threshold of contribution required too. The joke in the medical world is if you mention a paper over coffee, everybody at the table is a coauthor. Ecologists expect a bit more, which I think is good. But I’m going to argue here that ecologists have gone too far in the other direction by a (usually implicit) view that if you’re not substantially involved in the writing per se, you haven’t made an intellectual contribution. This has been the historical justification for excluding undergrads and technicians who do substantial data collection (which is nominally one of the core ways to contribute, but often not valued as much)*.

So let us examine who deserves authorship on a MAP (massively-authored-paper) that comes out of a working group against the above criteria (and look at the role the bias towards writing plays). Assume 15 people meet together for a week. Six months later a largely overlapping group meets for another week (maybe bringing the total involved up to 18). In the interim a subset of people (often more junior) do a lot of data analysis (again with overlap but maybe also in the PIs lab, bringing the total up to 20). Usually after the last meeting a couple of the leaders take the outline and figures developed as a group and turn it into a full paper. Who deserves authorship? Personally, I am a pretty strong believer in this answer: everybody who attended the working group or worked on the analysis deserves authorship, with the right to opt out if you don’t feel your contribution was large enough. The difference between opt out and opt in is subtle but important. In opt out your name is on the paper until you ask off. You are included in all the emails updating progress and your name is on the author page of the first draft. Opt-in means you might get the emails but you are not listed as an author until you raise your hand and ask to be on it. In my experience opt-in favors entitled senior faculty and disfavors early career folks who often make bigger contributions but might feel a certain imposter syndrome in raising their hand. So I prefer opt-out. And I think it is important that everybody who attends be in the opt-out list. The whole point of a working group is a free-flowing exchange of ideas without needing to track and assign credit. Putting some threshold of contribution breaks the dynamic (and again favors later career folks).

The other key issue of authorship (whether a two author paper or 20 author working-group-based opt-out paper) is that everybody has to agree to have their name on the final draft as authorship implies intellectual and ethical responsibility for the contents of the paper. If it is two authors this is easy – the two authors have probably gone back and forth editing each other until both are satisfied. But how does this work with 20 authors? It is impossible to write effectively with 20 people. A core group of 2-3 people really needs to do the writing in my experience. So then how do the remaining 17-18 people have some chance for input and for asserting ownership of the contents?

I can tell you what everybody does right now, and which I think is horribly inefficient. After the core group of 2-3 lead authors carefully polishes the document, it goes into a google doc and the other authors are invited to edit the google doc. And by and large the culture is if you don’t edit it, you are considered a slacker and maybe not valid as an author. This is horrible for a couple of reasons. First it produces writing that sounds like (because it is) committee writing. One person hates split infinitives. Another person likes the style to sound informal. A third adds a bunch of self-citations. A fourth really wants the introduction to start from a different point but doesn’t want to be that person who rewrites the whole introduction so stretches it to be neither fish nor fowl. And the document quickly becomes a sea of track changes nobody can even decipher. And then the poor lead authors have to try to cobble it back together into a coherent, stylistically unified document which is a LOT of work. It rarely makes the paper better. Revealingly, something like 90% of the comments are in the first half of the paper, and probably 50% in the first 10% of the paper. Few people have the stamina to make strong edits through to the end. And, again, it can be hard for an early career person to overwrite the ink of a senior person, even when the ECR has a better idea. And rarely are there major intellectual objections (presuming the lead authors were good listeners during the working group and are at least modestly diplomatic).  And quite often the lead authors have to do a good bit of rewriting to merge the diverse input, which will not be subject to review again. To my mind this approach exists because of the subtle bias that “writing” is a more important contribution than data collection or analysis or conceptualization of ideas. “If you haven’t written part of the paper you’re not really an author.”

I propose the following alternative. Recognize that contributing to any of conceptualization, data, analysis or writing (or funding) is a sufficient contribution. Agree in advance that participating in the working group or doing analysis automatically means you have made a sufficient contribution to qualify as an author. You were part of the free-flowing dialog that was the main value of a working group paper – the building of consensus and shared direction. And more prosaically, traveling for two weeks of your life is a much bigger commitment than a little half-hearted editing at a late stage. So you deserve authorship – that is no longer a question. Trust the leaders to honestly and accurately write-up the collectively envisioned paper (and if the group is split into two camps, make sure both camps are represented in the lead writing group). And now at the end forget about trying to determine who is an author (that really honestly should have already been decided by the end of the working group). Focus instead on the next step of ensuring ownership: making sure everybody who is already an author on the paper agrees to the version going out. Do this by:

This is a major change to the culture in ecology. And to be clear, I am not proposing it for 3-5 author traditional collaborations. But I believe it is the right direction to go for MAP/working-group papers. My proposal acknowledges that writing is not a contribution that trumps other contributions. It acknowledges the intellectual value of open ended discussion at a working group (as well as data munging). It provides cleaner, better writing. It preserves the sanity of the lead author (who is often early career and junior to the people who are making mostly stylistic corrections). And it provides a robust and legitimate path for authors to own their authorship on something much closer to the final draft than what the google doc route provides.

If you find yourself resisting my proposal, try the logical counter factual. Create a first draft of an ambitious new idea by yourself. Invite 20 people to edit it and use google docs to try to build a consensus. The intellectual contribution is now 100% through writing! How do you predict that will turn out? I’ve tried it a few times. It does not end well!** Those open ended discussions in person (including real time data analysis and results and interpretation) are why we have working groups, not shared editing***. So why not place the discussions at the center of qualifying for authorship. And make the poor lead author’s life easier, and the reader’s brain better able to focus on ideas not ever drifting stylistic conventions and awkward transitions.

What do you think? Do we overvalue writing as the primary intellectual contribution to a paper? Are there other approaches for MAPs that you’ve seen work well?

*And no I’m not saying every undergraduate who spends 20 hours in your lab grinding samples should be an author, but if they spent an entire summer collecting >1/3 of the data, maybe they should. And distinct from the point of this article, maybe they should be included at least a bit in the writing as a learning process.

**To be precise it does not end well with 10-20 authors. And it does not end well without prior in-person discussions about what the message of the paper is. The co-edit approach works great and is the preferred mode with a 2-5 person collaboration that has been meeting with each other through the development of the project. But I’m talking about working group MAPs here.

*** The increasing quality of LLM output is also going to force us to reckon with any biases that writing is the major intellectual contribution. They’re not there yet, but how long until an LLM could create transcripts of our meetings and then write a decent first draft of a paper? I bet that creative question development, method development, and analysis will remain the exclusive purview of human brains much longer.

Not an entirely surprising result but impressive to see empirically: “most GEA candidates did not exhibit GxE” in experimental validation of environmental associations in Arabidopsis

https://doi.org/10.1101/2025.01.08.631904

#science #genetics #Evolution